id	sid	tid	token	lemma	pos
app01-11088	1	1	acta	acta	PROPN
app01-11088	1	2	polytechnica	polytechnica	PROPN
app01-11088	1	3	ctu	ctu	NOUN
app01-11088	1	4	proceedings	proceeding	NOUN
app01-11088	1	5	https://doi.org/10.14311/app.2025.54.0017	https://doi.org/10.14311/app.2025.54.0017	PROPN
app01-11088	1	6	acta	acta	PROPN
app01-11088	1	7	polytechnica	polytechnica	PROPN
app01-11088	1	8	ctu	ctu	NOUN
app01-11088	1	9	proceedings	proceeding	NOUN
app01-11088	1	10	54:17–22	54:17–22	NUM
app01-11088	1	11	,	,	PUNCT
app01-11088	1	12	2025	2025	NUM
app01-11088	1	13	©	©	ADP
app01-11088	1	14	2025	2025	NUM
app01-11088	1	15	the	the	DET
app01-11088	1	16	author(s	author(s	NOUN
app01-11088	1	17	)	)	PUNCT
app01-11088	1	18	.	.	PUNCT
app01-11088	2	1	licensed	license	VERB
app01-11088	2	2	under	under	ADP
app01-11088	2	3	a	a	DET
app01-11088	2	4	cc	cc	NOUN
app01-11088	2	5	-	-	PUNCT
app01-11088	2	6	by	by	ADP
app01-11088	2	7	4.0	4.0	NUM
app01-11088	2	8	licence	licence	NOUN
app01-11088	2	9	published	publish	VERB
app01-11088	2	10	by	by	ADP
app01-11088	2	11	the	the	DET
app01-11088	2	12	czech	czech	PROPN
app01-11088	2	13	technical	technical	PROPN
app01-11088	2	14	university	university	PROPN
app01-11088	2	15	in	in	ADP
app01-11088	2	16	prague	prague	PROPN
app01-11088	2	17	tractable	tractable	ADJ
app01-11088	2	18	descriptors	descriptor	NOUN
app01-11088	2	19	for	for	ADP
app01-11088	2	20	digital	digital	ADJ
app01-11088	2	21	aggregate	aggregate	ADJ
app01-11088	2	22	generation	generation	NOUN
app01-11088	2	23	kaustav	kaustav	NOUN
app01-11088	2	24	das	das	PROPN
app01-11088	2	25	,	,	PUNCT
app01-11088	2	26	jan	jan	PROPN
app01-11088	2	27	sýkora∗	sýkora∗	PROPN
app01-11088	2	28	,	,	PUNCT
app01-11088	2	29	anna	anna	PROPN
app01-11088	2	30	kučerová	kučerová	AUX
app01-11088	2	31	czech	czech	PROPN
app01-11088	2	32	technical	technical	PROPN
app01-11088	2	33	university	university	PROPN
app01-11088	2	34	in	in	ADP
app01-11088	2	35	prague	prague	PROPN
app01-11088	2	36	,	,	PUNCT
app01-11088	2	37	faculty	faculty	NOUN
app01-11088	2	38	of	of	ADP
app01-11088	2	39	civil	civil	ADJ
app01-11088	2	40	engineering	engineering	NOUN
app01-11088	2	41	,	,	PUNCT
app01-11088	2	42	department	department	NOUN
app01-11088	2	43	of	of	ADP
app01-11088	2	44	mechanics	mechanic	NOUN
app01-11088	2	45	,	,	PUNCT
app01-11088	2	46	thákurova	thákurova	PROPN
app01-11088	2	47	2077/7	2077/7	NUM
app01-11088	2	48	,	,	PUNCT
app01-11088	2	49	160	160	NUM
app01-11088	2	50	00	00	NUM
app01-11088	2	51	prague	prague	PROPN
app01-11088	2	52	6	6	NUM
app01-11088	2	53	–	–	PUNCT
app01-11088	2	54	dejvice	dejvice	NOUN
app01-11088	2	55	,	,	PUNCT
app01-11088	2	56	czech	czech	PROPN
app01-11088	2	57	republic	republic	NOUN
app01-11088	2	58	∗	∗	NOUN
app01-11088	2	59	corresponding	correspond	VERB
app01-11088	2	60	author	author	NOUN
app01-11088	2	61	:	:	PUNCT
app01-11088	2	62	jan.sykora.1@cvut.cz	jan.sykora.1@cvut.cz	NOUN
app01-11088	2	63	abstract	abstract	ADJ
app01-11088	2	64	.	.	PUNCT
app01-11088	3	1	this	this	DET
app01-11088	3	2	paper	paper	NOUN
app01-11088	3	3	introduces	introduce	VERB
app01-11088	3	4	a	a	DET
app01-11088	3	5	deep	deep	ADJ
app01-11088	3	6	learning	learning	NOUN
app01-11088	3	7	-	-	PUNCT
app01-11088	3	8	based	base	VERB
app01-11088	3	9	approach	approach	NOUN
app01-11088	3	10	for	for	ADP
app01-11088	3	11	generating	generate	VERB
app01-11088	3	12	2d	2d	NUM
app01-11088	3	13	aggregate	aggregate	ADJ
app01-11088	3	14	shapes	shape	NOUN
app01-11088	3	15	using	use	VERB
app01-11088	3	16	a	a	DET
app01-11088	3	17	small	small	ADJ
app01-11088	3	18	set	set	NOUN
app01-11088	3	19	of	of	ADP
app01-11088	3	20	tractable	tractable	ADJ
app01-11088	3	21	and	and	CCONJ
app01-11088	3	22	physically	physically	ADV
app01-11088	3	23	meaningful	meaningful	ADJ
app01-11088	3	24	parameters	parameter	NOUN
app01-11088	3	25	.	.	PUNCT
app01-11088	4	1	in	in	ADP
app01-11088	4	2	contrast	contrast	NOUN
app01-11088	4	3	to	to	ADP
app01-11088	4	4	existing	exist	VERB
app01-11088	4	5	methods	method	NOUN
app01-11088	4	6	that	that	PRON
app01-11088	4	7	often	often	ADV
app01-11088	4	8	rely	rely	VERB
app01-11088	4	9	on	on	ADP
app01-11088	4	10	quantities	quantity	NOUN
app01-11088	4	11	without	without	ADP
app01-11088	4	12	clear	clear	ADJ
app01-11088	4	13	physical	physical	ADJ
app01-11088	4	14	interpretations	interpretation	NOUN
app01-11088	4	15	,	,	PUNCT
app01-11088	4	16	the	the	DET
app01-11088	4	17	approach	approach	NOUN
app01-11088	4	18	presented	present	VERB
app01-11088	4	19	here	here	ADV
app01-11088	4	20	leverages	leverage	NOUN
app01-11088	4	21	scale	scale	NOUN
app01-11088	4	22	invariant	invariant	ADJ
app01-11088	4	23	moments	moment	NOUN
app01-11088	4	24	to	to	PART
app01-11088	4	25	capture	capture	VERB
app01-11088	4	26	essential	essential	ADJ
app01-11088	4	27	shape	shape	NOUN
app01-11088	4	28	characteristics	characteristic	NOUN
app01-11088	4	29	.	.	PUNCT
app01-11088	5	1	additionally	additionally	ADV
app01-11088	5	2	,	,	PUNCT
app01-11088	5	3	we	we	PRON
app01-11088	5	4	introduce	introduce	VERB
app01-11088	5	5	the	the	DET
app01-11088	5	6	idea	idea	NOUN
app01-11088	5	7	of	of	ADP
app01-11088	5	8	using	use	VERB
app01-11088	5	9	scale	scale	NOUN
app01-11088	5	10	-	-	PUNCT
app01-11088	5	11	invariant	invariant	ADJ
app01-11088	5	12	state	state	NOUN
app01-11088	5	13	descriptors	descriptor	NOUN
app01-11088	5	14	,	,	PUNCT
app01-11088	5	15	such	such	ADJ
app01-11088	5	16	as	as	ADP
app01-11088	5	17	area	area	NOUN
app01-11088	5	18	,	,	PUNCT
app01-11088	5	19	to	to	PART
app01-11088	5	20	control	control	VERB
app01-11088	5	21	the	the	DET
app01-11088	5	22	size	size	NOUN
app01-11088	5	23	of	of	ADP
app01-11088	5	24	the	the	DET
app01-11088	5	25	generated	generate	VERB
app01-11088	5	26	shapes	shape	NOUN
app01-11088	5	27	.	.	PUNCT
app01-11088	6	1	the	the	DET
app01-11088	6	2	neural	neural	ADJ
app01-11088	6	3	network	network	NOUN
app01-11088	6	4	is	be	AUX
app01-11088	6	5	trained	train	VERB
app01-11088	6	6	to	to	PART
app01-11088	6	7	generate	generate	VERB
app01-11088	6	8	shapes	shape	NOUN
app01-11088	6	9	corresponding	correspond	VERB
app01-11088	6	10	to	to	ADP
app01-11088	6	11	these	these	DET
app01-11088	6	12	parameters	parameter	NOUN
app01-11088	6	13	,	,	PUNCT
app01-11088	6	14	and	and	CCONJ
app01-11088	6	15	its	its	PRON
app01-11088	6	16	ability	ability	NOUN
app01-11088	6	17	to	to	PART
app01-11088	6	18	learn	learn	VERB
app01-11088	6	19	the	the	DET
app01-11088	6	20	relationship	relationship	NOUN
app01-11088	6	21	between	between	ADP
app01-11088	6	22	shape	shape	NOUN
app01-11088	6	23	constants	constant	NOUN
app01-11088	6	24	and	and	CCONJ
app01-11088	6	25	state	state	NOUN
app01-11088	6	26	descriptors	descriptor	NOUN
app01-11088	6	27	without	without	ADP
app01-11088	6	28	explicit	explicit	ADJ
app01-11088	6	29	data	datum	NOUN
app01-11088	6	30	augmentation	augmentation	NOUN
app01-11088	6	31	is	be	AUX
app01-11088	6	32	demonstrated	demonstrate	VERB
app01-11088	6	33	.	.	PUNCT
app01-11088	7	1	the	the	DET
app01-11088	7	2	framework	framework	NOUN
app01-11088	7	3	thus	thus	ADV
app01-11088	7	4	presented	present	VERB
app01-11088	7	5	provides	provide	VERB
app01-11088	7	6	a	a	DET
app01-11088	7	7	foundation	foundation	NOUN
app01-11088	7	8	for	for	ADP
app01-11088	7	9	developing	develop	VERB
app01-11088	7	10	microstructure	microstructure	ADJ
app01-11088	7	11	generators	generator	NOUN
app01-11088	7	12	that	that	PRON
app01-11088	7	13	offer	offer	VERB
app01-11088	7	14	enhanced	enhance	VERB
app01-11088	7	15	interpretability	interpretability	NOUN
app01-11088	7	16	by	by	ADP
app01-11088	7	17	relying	rely	VERB
app01-11088	7	18	on	on	ADP
app01-11088	7	19	parameters	parameter	NOUN
app01-11088	7	20	that	that	PRON
app01-11088	7	21	provide	provide	VERB
app01-11088	7	22	meaningful	meaningful	ADJ
app01-11088	7	23	insights	insight	NOUN
app01-11088	7	24	into	into	ADP
app01-11088	7	25	the	the	DET
app01-11088	7	26	description	description	NOUN
app01-11088	7	27	of	of	ADP
app01-11088	7	28	material	material	NOUN
app01-11088	7	29	morphology	morphology	NOUN
app01-11088	7	30	composed	compose	VERB
app01-11088	7	31	of	of	ADP
app01-11088	7	32	non	non	ADJ
app01-11088	7	33	-	-	ADJ
app01-11088	7	34	trivial	trivial	ADJ
app01-11088	7	35	shapes	shape	NOUN
app01-11088	7	36	.	.	PUNCT
app01-11088	8	1	keywords	keyword	NOUN
app01-11088	8	2	:	:	PUNCT
app01-11088	8	3	reconstruction	reconstruction	NOUN
app01-11088	8	4	,	,	PUNCT
app01-11088	8	5	moment	moment	NOUN
app01-11088	8	6	invariants	invariant	NOUN
app01-11088	8	7	,	,	PUNCT
app01-11088	8	8	deep	deep	ADJ
app01-11088	8	9	learning	learning	NOUN
app01-11088	8	10	,	,	PUNCT
app01-11088	8	11	concrete	concrete	ADJ
app01-11088	8	12	.	.	PUNCT
app01-11088	9	1	1	1	X
app01-11088	9	2	.	.	X
app01-11088	9	3	introduction	introduction	NOUN
app01-11088	9	4	concrete	concrete	NOUN
app01-11088	9	5	is	be	AUX
app01-11088	9	6	primarily	primarily	ADV
app01-11088	9	7	a	a	DET
app01-11088	9	8	matrix	matrix	NOUN
app01-11088	9	9	inclusion	inclusion	NOUN
app01-11088	9	10	composite	composite	NOUN
app01-11088	9	11	consisting	consist	VERB
app01-11088	9	12	of	of	ADP
app01-11088	9	13	a	a	DET
app01-11088	9	14	mortar	mortar	NOUN
app01-11088	9	15	-	-	PUNCT
app01-11088	9	16	based	base	VERB
app01-11088	9	17	matrix	matrix	NOUN
app01-11088	9	18	embedded	embed	VERB
app01-11088	9	19	with	with	ADP
app01-11088	9	20	aggregate	aggregate	NOUN
app01-11088	9	21	-	-	PUNCT
app01-11088	9	22	based	base	VERB
app01-11088	9	23	inclusions	inclusion	NOUN
app01-11088	9	24	.	.	PUNCT
app01-11088	10	1	since	since	SCONJ
app01-11088	10	2	the	the	DET
app01-11088	10	3	casting	cast	VERB
app01-11088	10	4	process	process	NOUN
app01-11088	10	5	results	result	VERB
app01-11088	10	6	in	in	ADP
app01-11088	10	7	the	the	DET
app01-11088	10	8	introduction	introduction	NOUN
app01-11088	10	9	of	of	ADP
app01-11088	10	10	porosity	porosity	NOUN
app01-11088	10	11	defects	defect	NOUN
app01-11088	10	12	and	and	CCONJ
app01-11088	10	13	also	also	ADV
app01-11088	10	14	leads	lead	VERB
app01-11088	10	15	to	to	ADP
app01-11088	10	16	the	the	DET
app01-11088	10	17	formation	formation	NOUN
app01-11088	10	18	of	of	ADP
app01-11088	10	19	interfacial	interfacial	ADJ
app01-11088	10	20	transition	transition	NOUN
app01-11088	10	21	zones	zone	NOUN
app01-11088	10	22	(	(	PUNCT
app01-11088	10	23	itz	itz	PROPN
app01-11088	10	24	)	)	PUNCT
app01-11088	10	25	in	in	ADP
app01-11088	10	26	the	the	DET
app01-11088	10	27	regions	region	NOUN
app01-11088	10	28	between	between	ADP
app01-11088	10	29	the	the	DET
app01-11088	10	30	aggregates	aggregate	NOUN
app01-11088	10	31	and	and	CCONJ
app01-11088	10	32	the	the	DET
app01-11088	10	33	surrounding	surround	VERB
app01-11088	10	34	mortar	mortar	NOUN
app01-11088	10	35	,	,	PUNCT
app01-11088	10	36	these	these	PRON
app01-11088	10	37	are	be	AUX
app01-11088	10	38	also	also	ADV
app01-11088	10	39	treated	treat	VERB
app01-11088	10	40	as	as	ADP
app01-11088	10	41	additional	additional	ADJ
app01-11088	10	42	phases	phase	NOUN
app01-11088	10	43	during	during	ADP
app01-11088	10	44	analysis	analysis	NOUN
app01-11088	10	45	[	[	X
app01-11088	10	46	1	1	NUM
app01-11088	10	47	,	,	PUNCT
app01-11088	10	48	2	2	NUM
app01-11088	10	49	]	]	PUNCT
app01-11088	10	50	.	.	PUNCT
app01-11088	11	1	the	the	DET
app01-11088	11	2	heterogeneous	heterogeneous	ADJ
app01-11088	11	3	structure	structure	NOUN
app01-11088	11	4	formed	form	VERB
app01-11088	11	5	by	by	ADP
app01-11088	11	6	the	the	DET
app01-11088	11	7	spatial	spatial	ADJ
app01-11088	11	8	arrangement	arrangement	NOUN
app01-11088	11	9	of	of	ADP
app01-11088	11	10	these	these	DET
app01-11088	11	11	constitutive	constitutive	ADJ
app01-11088	11	12	phases	phase	NOUN
app01-11088	11	13	is	be	AUX
app01-11088	11	14	herein	herein	NOUN
app01-11088	11	15	termed	term	VERB
app01-11088	11	16	the	the	DET
app01-11088	11	17	concrete	concrete	ADJ
app01-11088	11	18	microstructure	microstructure	NOUN
app01-11088	11	19	.	.	PUNCT
app01-11088	12	1	the	the	DET
app01-11088	12	2	role	role	NOUN
app01-11088	12	3	of	of	ADP
app01-11088	12	4	concrete	concrete	ADJ
app01-11088	12	5	microstructure	microstructure	NOUN
app01-11088	12	6	on	on	ADP
app01-11088	12	7	its	its	PRON
app01-11088	12	8	overall	overall	ADJ
app01-11088	12	9	performance	performance	NOUN
app01-11088	12	10	is	be	AUX
app01-11088	12	11	well	well	ADV
app01-11088	12	12	established	establish	VERB
app01-11088	12	13	with	with	ADP
app01-11088	12	14	correlation	correlation	NOUN
app01-11088	12	15	being	be	AUX
app01-11088	12	16	observed	observe	VERB
app01-11088	12	17	in	in	ADP
app01-11088	12	18	its	its	PRON
app01-11088	12	19	mechanical	mechanical	ADJ
app01-11088	12	20	behaviour	behaviour	NOUN
app01-11088	12	21	[	[	X
app01-11088	12	22	3	3	NUM
app01-11088	12	23	]	]	PUNCT
app01-11088	12	24	,	,	PUNCT
app01-11088	12	25	crack	crack	VERB
app01-11088	12	26	formation	formation	NOUN
app01-11088	12	27	[	[	X
app01-11088	12	28	4	4	NUM
app01-11088	12	29	]	]	PUNCT
app01-11088	12	30	,	,	PUNCT
app01-11088	12	31	hydration	hydration	NOUN
app01-11088	12	32	mechanisms	mechanism	NOUN
app01-11088	12	33	[	[	X
app01-11088	12	34	5	5	NUM
app01-11088	12	35	]	]	PUNCT
app01-11088	12	36	,	,	PUNCT
app01-11088	12	37	thermo	thermo	NOUN
app01-11088	12	38	-	-	PUNCT
app01-11088	12	39	mechanical	mechanical	ADJ
app01-11088	12	40	coupling	coupling	NOUN
app01-11088	13	1	[	[	X
app01-11088	13	2	6	6	NUM
app01-11088	13	3	]	]	PUNCT
app01-11088	13	4	to	to	PART
app01-11088	13	5	name	name	VERB
app01-11088	13	6	a	a	DET
app01-11088	13	7	few	few	ADJ
app01-11088	13	8	.	.	PUNCT
app01-11088	14	1	in	in	ADP
app01-11088	14	2	particular	particular	ADJ
app01-11088	14	3	,	,	PUNCT
app01-11088	14	4	the	the	DET
app01-11088	14	5	size	size	NOUN
app01-11088	14	6	,	,	PUNCT
app01-11088	14	7	shape	shape	NOUN
app01-11088	14	8	and	and	CCONJ
app01-11088	14	9	the	the	DET
app01-11088	14	10	roughness	roughness	NOUN
app01-11088	14	11	of	of	ADP
app01-11088	14	12	aggregates	aggregate	NOUN
app01-11088	14	13	play	play	VERB
app01-11088	14	14	an	an	DET
app01-11088	14	15	important	important	ADJ
app01-11088	14	16	role	role	NOUN
app01-11088	14	17	in	in	ADP
app01-11088	14	18	the	the	DET
app01-11088	14	19	overall	overall	ADJ
app01-11088	14	20	strength	strength	NOUN
app01-11088	14	21	of	of	ADP
app01-11088	14	22	concretes	concrete	NOUN
app01-11088	14	23	[	[	X
app01-11088	14	24	7	7	NUM
app01-11088	14	25	]	]	PUNCT
app01-11088	14	26	and	and	CCONJ
app01-11088	14	27	the	the	DET
app01-11088	14	28	development	development	NOUN
app01-11088	14	29	of	of	ADP
app01-11088	14	30	itzs	itz	NOUN
app01-11088	14	31	[	[	X
app01-11088	14	32	8	8	NUM
app01-11088	14	33	]	]	PUNCT
app01-11088	14	34	,	,	PUNCT
app01-11088	14	35	which	which	PRON
app01-11088	14	36	are	be	AUX
app01-11088	14	37	accepted	accept	VERB
app01-11088	14	38	as	as	ADP
app01-11088	14	39	the	the	DET
app01-11088	14	40	weakest	weak	ADJ
app01-11088	14	41	phase	phase	NOUN
app01-11088	14	42	in	in	ADP
app01-11088	14	43	concretes	concrete	NOUN
app01-11088	14	44	[	[	X
app01-11088	14	45	9	9	NUM
app01-11088	14	46	]	]	PUNCT
app01-11088	14	47	.	.	PUNCT
app01-11088	15	1	however	however	ADV
app01-11088	15	2	,	,	PUNCT
app01-11088	15	3	the	the	DET
app01-11088	15	4	grading	grading	NOUN
app01-11088	15	5	of	of	ADP
app01-11088	15	6	aggregates	aggregate	NOUN
app01-11088	15	7	,	,	PUNCT
app01-11088	15	8	their	their	PRON
app01-11088	15	9	property	property	NOUN
app01-11088	15	10	,	,	PUNCT
app01-11088	15	11	and	and	CCONJ
app01-11088	15	12	their	their	PRON
app01-11088	15	13	type	type	NOUN
app01-11088	15	14	are	be	AUX
app01-11088	15	15	also	also	ADV
app01-11088	15	16	among	among	ADP
app01-11088	15	17	the	the	DET
app01-11088	15	18	controllable	controllable	ADJ
app01-11088	15	19	parameters	parameter	NOUN
app01-11088	15	20	in	in	ADP
app01-11088	15	21	concrete	concrete	ADJ
app01-11088	15	22	mixtures	mixture	NOUN
app01-11088	15	23	.	.	PUNCT
app01-11088	16	1	therefore	therefore	ADV
app01-11088	16	2	,	,	PUNCT
app01-11088	16	3	due	due	ADP
app01-11088	16	4	to	to	ADP
app01-11088	16	5	the	the	DET
app01-11088	16	6	vast	vast	ADJ
app01-11088	16	7	number	number	NOUN
app01-11088	16	8	of	of	ADP
app01-11088	16	9	possible	possible	ADJ
app01-11088	16	10	combinations	combination	NOUN
app01-11088	16	11	of	of	ADP
app01-11088	16	12	parameters	parameter	NOUN
app01-11088	16	13	,	,	PUNCT
app01-11088	16	14	development	development	NOUN
app01-11088	16	15	of	of	ADP
app01-11088	16	16	concrete	concrete	ADJ
app01-11088	16	17	mixtures	mixture	NOUN
app01-11088	16	18	for	for	ADP
app01-11088	16	19	specific	specific	ADJ
app01-11088	16	20	applications	application	NOUN
app01-11088	16	21	requires	require	VERB
app01-11088	16	22	substantial	substantial	ADJ
app01-11088	16	23	investment	investment	NOUN
app01-11088	16	24	of	of	ADP
app01-11088	16	25	time	time	NOUN
app01-11088	16	26	and	and	CCONJ
app01-11088	16	27	resources	resource	NOUN
app01-11088	16	28	if	if	SCONJ
app01-11088	16	29	performed	perform	VERB
app01-11088	16	30	by	by	ADP
app01-11088	16	31	exclusively	exclusively	ADV
app01-11088	16	32	employing	employ	VERB
app01-11088	16	33	experimental	experimental	ADJ
app01-11088	16	34	methods	method	NOUN
app01-11088	16	35	.	.	PUNCT
app01-11088	17	1	alternatively	alternatively	ADV
app01-11088	17	2	,	,	PUNCT
app01-11088	17	3	numerical	numerical	ADJ
app01-11088	17	4	schemes	scheme	NOUN
app01-11088	17	5	enable	enable	VERB
app01-11088	17	6	an	an	DET
app01-11088	17	7	effective	effective	ADJ
app01-11088	17	8	preliminary	preliminary	ADJ
app01-11088	17	9	study	study	NOUN
app01-11088	17	10	of	of	ADP
app01-11088	17	11	concretes	concrete	NOUN
app01-11088	17	12	and	and	CCONJ
app01-11088	17	13	can	can	AUX
app01-11088	17	14	greatly	greatly	ADV
app01-11088	17	15	reduce	reduce	VERB
app01-11088	17	16	experiments	experiment	NOUN
app01-11088	17	17	by	by	ADP
app01-11088	17	18	facilitating	facilitate	VERB
app01-11088	17	19	the	the	DET
app01-11088	17	20	inspection	inspection	NOUN
app01-11088	17	21	of	of	ADP
app01-11088	17	22	a	a	DET
app01-11088	17	23	larger	large	ADJ
app01-11088	17	24	set	set	NOUN
app01-11088	17	25	of	of	ADP
app01-11088	17	26	parameters	parameter	NOUN
app01-11088	17	27	in	in	ADP
app01-11088	17	28	-	-	PUNCT
app01-11088	17	29	silico	silico	NOUN
app01-11088	17	30	[	[	X
app01-11088	17	31	2	2	NUM
app01-11088	17	32	]	]	PUNCT
app01-11088	17	33	.	.	PUNCT
app01-11088	18	1	for	for	ADP
app01-11088	18	2	materials	material	NOUN
app01-11088	18	3	such	such	ADJ
app01-11088	18	4	as	as	ADP
app01-11088	18	5	concretes	concrete	NOUN
app01-11088	18	6	where	where	SCONJ
app01-11088	18	7	scale	scale	NOUN
app01-11088	18	8	separation	separation	NOUN
app01-11088	18	9	is	be	AUX
app01-11088	18	10	well	well	ADV
app01-11088	18	11	defined	define	VERB
app01-11088	18	12	,	,	PUNCT
app01-11088	18	13	multi	multi	ADJ
app01-11088	18	14	-	-	ADJ
app01-11088	18	15	scale	scale	ADJ
app01-11088	18	16	homogenization	homogenization	NOUN
app01-11088	18	17	schemes	scheme	NOUN
app01-11088	18	18	are	be	AUX
app01-11088	18	19	used	use	VERB
app01-11088	18	20	for	for	ADP
app01-11088	18	21	determining	determine	VERB
app01-11088	18	22	the	the	DET
app01-11088	18	23	bulk	bulk	ADJ
app01-11088	18	24	response	response	NOUN
app01-11088	18	25	by	by	ADP
app01-11088	18	26	taking	take	VERB
app01-11088	18	27	into	into	ADP
app01-11088	18	28	consideration	consideration	NOUN
app01-11088	18	29	material	material	NOUN
app01-11088	18	30	heterogeneity	heterogeneity	NOUN
app01-11088	18	31	[	[	X
app01-11088	18	32	10	10	NUM
app01-11088	18	33	]	]	PUNCT
app01-11088	18	34	.	.	PUNCT
app01-11088	19	1	however	however	ADV
app01-11088	19	2	,	,	PUNCT
app01-11088	19	3	accuracy	accuracy	NOUN
app01-11088	19	4	of	of	ADP
app01-11088	19	5	numerical	numerical	ADJ
app01-11088	19	6	schemes	scheme	NOUN
app01-11088	19	7	depend	depend	VERB
app01-11088	19	8	upon	upon	SCONJ
app01-11088	19	9	the	the	DET
app01-11088	19	10	level	level	NOUN
app01-11088	19	11	of	of	ADP
app01-11088	19	12	fidelity	fidelity	NOUN
app01-11088	19	13	used	use	VERB
app01-11088	19	14	for	for	ADP
app01-11088	19	15	the	the	DET
app01-11088	19	16	definition	definition	NOUN
app01-11088	19	17	of	of	ADP
app01-11088	19	18	the	the	DET
app01-11088	19	19	computational	computational	ADJ
app01-11088	19	20	domain	domain	NOUN
app01-11088	19	21	which	which	PRON
app01-11088	19	22	in	in	ADP
app01-11088	19	23	this	this	DET
app01-11088	19	24	case	case	NOUN
app01-11088	19	25	,	,	PUNCT
app01-11088	19	26	is	be	AUX
app01-11088	19	27	linked	link	VERB
app01-11088	19	28	to	to	ADP
app01-11088	19	29	the	the	DET
app01-11088	19	30	distribution	distribution	NOUN
app01-11088	19	31	of	of	ADP
app01-11088	19	32	aggregates	aggregate	NOUN
app01-11088	19	33	.	.	PUNCT
app01-11088	20	1	generation	generation	NOUN
app01-11088	20	2	of	of	ADP
app01-11088	20	3	complex	complex	ADJ
app01-11088	20	4	computational	computational	ADJ
app01-11088	20	5	domains	domain	NOUN
app01-11088	20	6	or	or	CCONJ
app01-11088	20	7	microstructures	microstructure	NOUN
app01-11088	20	8	fall	fall	VERB
app01-11088	20	9	into	into	ADP
app01-11088	20	10	the	the	DET
app01-11088	20	11	purview	purview	NOUN
app01-11088	20	12	of	of	ADP
app01-11088	20	13	microstructure	microstructure	ADJ
app01-11088	20	14	characterization	characterization	NOUN
app01-11088	20	15	and	and	CCONJ
app01-11088	20	16	reconstruction	reconstruction	NOUN
app01-11088	20	17	(	(	PUNCT
app01-11088	20	18	mcr	mcr	PROPN
app01-11088	20	19	)	)	PUNCT
app01-11088	20	20	techniques	technique	NOUN
app01-11088	20	21	where	where	SCONJ
app01-11088	20	22	some	some	PRON
app01-11088	20	23	of	of	ADP
app01-11088	20	24	the	the	DET
app01-11088	20	25	widely	widely	ADV
app01-11088	20	26	adopted	adopt	VERB
app01-11088	20	27	methods	method	NOUN
app01-11088	20	28	for	for	ADP
app01-11088	20	29	modeling	model	VERB
app01-11088	20	30	heterogeneous	heterogeneous	ADJ
app01-11088	20	31	material	material	NOUN
app01-11088	20	32	structures	structure	NOUN
app01-11088	20	33	either	either	CCONJ
app01-11088	20	34	include	include	VERB
app01-11088	20	35	the	the	DET
app01-11088	20	36	use	use	NOUN
app01-11088	20	37	of	of	ADP
app01-11088	20	38	very	very	ADV
app01-11088	20	39	large	large	ADJ
app01-11088	20	40	set	set	NOUN
app01-11088	20	41	of	of	ADP
app01-11088	20	42	parameters	parameter	NOUN
app01-11088	20	43	or	or	CCONJ
app01-11088	20	44	include	include	VERB
app01-11088	20	45	the	the	DET
app01-11088	20	46	use	use	NOUN
app01-11088	20	47	of	of	ADP
app01-11088	20	48	specialized	specialized	ADJ
app01-11088	20	49	functions	function	NOUN
app01-11088	20	50	without	without	ADP
app01-11088	20	51	well	well	ADV
app01-11088	20	52	defined	define	VERB
app01-11088	20	53	physical	physical	ADJ
app01-11088	20	54	meaning	meaning	NOUN
app01-11088	20	55	[	[	X
app01-11088	20	56	11	11	NUM
app01-11088	20	57	]	]	PUNCT
app01-11088	20	58	.	.	PUNCT
app01-11088	21	1	this	this	PRON
app01-11088	21	2	is	be	AUX
app01-11088	21	3	also	also	ADV
app01-11088	21	4	evident	evident	ADJ
app01-11088	21	5	in	in	ADP
app01-11088	21	6	the	the	DET
app01-11088	21	7	modeling	modeling	NOUN
app01-11088	21	8	of	of	ADP
app01-11088	21	9	aggregates	aggregate	NOUN
app01-11088	21	10	where	where	SCONJ
app01-11088	21	11	spherical	spherical	ADJ
app01-11088	21	12	harmonics	harmonic	NOUN
app01-11088	21	13	and	and	CCONJ
app01-11088	21	14	random	random	ADJ
app01-11088	21	15	fields	field	NOUN
app01-11088	21	16	see	see	VERB
app01-11088	21	17	wide	wide	ADJ
app01-11088	21	18	adoption	adoption	NOUN
app01-11088	21	19	[	[	X
app01-11088	21	20	12	12	NUM
app01-11088	21	21	]	]	PUNCT
app01-11088	21	22	,	,	PUNCT
app01-11088	21	23	which	which	PRON
app01-11088	21	24	by	by	ADP
app01-11088	21	25	themselves	themselves	PRON
app01-11088	21	26	lack	lack	VERB
app01-11088	21	27	clear	clear	ADJ
app01-11088	21	28	physical	physical	ADJ
app01-11088	21	29	meanings	meaning	NOUN
app01-11088	21	30	.	.	PUNCT
app01-11088	22	1	although	although	SCONJ
app01-11088	22	2	it	it	PRON
app01-11088	22	3	is	be	AUX
app01-11088	22	4	possible	possible	ADJ
app01-11088	22	5	to	to	PART
app01-11088	22	6	obtain	obtain	VERB
app01-11088	22	7	a	a	DET
app01-11088	22	8	limited	limited	ADJ
app01-11088	22	9	number	number	NOUN
app01-11088	22	10	of	of	ADP
app01-11088	22	11	physical	physical	ADJ
app01-11088	22	12	descriptors	descriptor	NOUN
app01-11088	22	13	based	base	VERB
app01-11088	22	14	on	on	ADP
app01-11088	22	15	polygonal	polygonal	ADJ
app01-11088	22	16	fits	fit	NOUN
app01-11088	22	17	such	such	ADJ
app01-11088	22	18	as	as	ADP
app01-11088	22	19	bounding	bounding	NOUN
app01-11088	22	20	boxes	box	NOUN
app01-11088	22	21	,	,	PUNCT
app01-11088	22	22	circumscribed	circumscribed	ADJ
app01-11088	22	23	circles	circle	NOUN
app01-11088	22	24	or	or	CCONJ
app01-11088	22	25	ellipses	ellipsis	NOUN
app01-11088	22	26	etc	etc	X
app01-11088	22	27	,	,	PUNCT
app01-11088	22	28	they	they	PRON
app01-11088	22	29	are	be	AUX
app01-11088	22	30	rarely	rarely	ADV
app01-11088	22	31	used	use	VERB
app01-11088	22	32	for	for	ADP
app01-11088	22	33	generation	generation	NOUN
app01-11088	22	34	of	of	ADP
app01-11088	22	35	aggregates	aggregate	NOUN
app01-11088	22	36	but	but	CCONJ
app01-11088	22	37	are	be	AUX
app01-11088	22	38	rather	rather	ADV
app01-11088	22	39	used	use	VERB
app01-11088	22	40	for	for	ADP
app01-11088	22	41	general	general	ADJ
app01-11088	22	42	evaluation	evaluation	NOUN
app01-11088	23	1	[	[	X
app01-11088	23	2	12	12	NUM
app01-11088	23	3	]	]	PUNCT
app01-11088	23	4	.	.	PUNCT
app01-11088	24	1	this	this	DET
app01-11088	24	2	paper	paper	NOUN
app01-11088	24	3	focuses	focus	VERB
app01-11088	24	4	on	on	ADP
app01-11088	24	5	the	the	DET
app01-11088	24	6	utilization	utilization	NOUN
app01-11088	24	7	of	of	ADP
app01-11088	24	8	deeplearning	deeplearne	VERB
app01-11088	24	9	(	(	PUNCT
app01-11088	24	10	dl	dl	NOUN
app01-11088	24	11	)	)	PUNCT
app01-11088	24	12	tool	tool	NOUN
app01-11088	24	13	for	for	ADP
app01-11088	24	14	generation	generation	NOUN
app01-11088	24	15	of	of	ADP
app01-11088	24	16	aggregate	aggregate	ADJ
app01-11088	24	17	shapes	shape	NOUN
app01-11088	24	18	in	in	ADP
app01-11088	24	19	2d	2d	NUM
app01-11088	24	20	by	by	ADP
app01-11088	24	21	utilizing	utilize	VERB
app01-11088	24	22	a	a	DET
app01-11088	24	23	small	small	ADJ
app01-11088	24	24	set	set	NOUN
app01-11088	24	25	of	of	ADP
app01-11088	24	26	tractable	tractable	ADJ
app01-11088	24	27	and	and	CCONJ
app01-11088	24	28	physically	physically	ADV
app01-11088	24	29	meaningful	meaningful	ADJ
app01-11088	24	30	parameters	parameter	NOUN
app01-11088	24	31	,	,	PUNCT
app01-11088	24	32	which	which	PRON
app01-11088	24	33	are	be	AUX
app01-11088	24	34	henceforth	henceforth	ADV
app01-11088	24	35	termed	term	VERB
app01-11088	24	36	as	as	ADP
app01-11088	24	37	geometric	geometric	ADJ
app01-11088	24	38	descriptors	descriptor	NOUN
app01-11088	24	39	[	[	X
app01-11088	24	40	13	13	NUM
app01-11088	24	41	]	]	PUNCT
app01-11088	24	42	of	of	ADP
app01-11088	24	43	an	an	DET
app01-11088	24	44	aggregate	aggregate	NOUN
app01-11088	24	45	.	.	PUNCT
app01-11088	25	1	in	in	ADP
app01-11088	25	2	this	this	DET
app01-11088	25	3	effect	effect	NOUN
app01-11088	25	4	,	,	PUNCT
app01-11088	25	5	the	the	DET
app01-11088	25	6	idea	idea	NOUN
app01-11088	25	7	of	of	ADP
app01-11088	25	8	moment	moment	NOUN
app01-11088	25	9	invaraints	invaraint	NOUN
app01-11088	25	10	for	for	ADP
app01-11088	25	11	image	image	NOUN
app01-11088	25	12	recognition	recognition	NOUN
app01-11088	25	13	is	be	AUX
app01-11088	25	14	revisited	revisit	VERB
app01-11088	25	15	which	which	PRON
app01-11088	25	16	was	be	AUX
app01-11088	25	17	originally	originally	ADV
app01-11088	25	18	proposed	propose	VERB
app01-11088	25	19	in	in	ADP
app01-11088	25	20	[	[	X
app01-11088	25	21	14	14	NUM
app01-11088	25	22	]	]	PUNCT
app01-11088	25	23	and	and	CCONJ
app01-11088	25	24	improved	improve	VERB
app01-11088	25	25	upon	upon	SCONJ
app01-11088	25	26	in	in	ADP
app01-11088	25	27	[	[	X
app01-11088	25	28	15	15	NUM
app01-11088	25	29	]	]	PUNCT
app01-11088	25	30	.	.	PUNCT
app01-11088	26	1	for	for	ADP
app01-11088	26	2	more	more	ADJ
app01-11088	26	3	information	information	NOUN
app01-11088	26	4	on	on	ADP
app01-11088	26	5	image	image	NOUN
app01-11088	26	6	moments	moment	NOUN
app01-11088	26	7	,	,	PUNCT
app01-11088	26	8	one	one	PRON
app01-11088	26	9	may	may	AUX
app01-11088	26	10	refer	refer	VERB
app01-11088	26	11	to	to	ADP
app01-11088	26	12	[	[	X
app01-11088	26	13	16	16	NUM
app01-11088	26	14	]	]	PUNCT
app01-11088	26	15	.	.	PUNCT
app01-11088	27	1	a	a	DET
app01-11088	27	2	key	key	ADJ
app01-11088	27	3	highlight	highlight	NOUN
app01-11088	27	4	in	in	ADP
app01-11088	27	5	these	these	DET
app01-11088	27	6	works	work	NOUN
app01-11088	27	7	is	be	AUX
app01-11088	27	8	in	in	ADP
app01-11088	27	9	the	the	DET
app01-11088	27	10	use	use	NOUN
app01-11088	27	11	of	of	ADP
app01-11088	27	12	geometric	geometric	ADJ
app01-11088	27	13	descriptors	descriptor	NOUN
app01-11088	27	14	that	that	PRON
app01-11088	27	15	are	be	AUX
app01-11088	27	16	shape	shape	NOUN
app01-11088	27	17	specific	specific	ADJ
app01-11088	27	18	and	and	CCONJ
app01-11088	27	19	do	do	AUX
app01-11088	27	20	not	not	PART
app01-11088	27	21	change	change	VERB
app01-11088	27	22	under	under	ADP
app01-11088	27	23	certain	certain	ADJ
app01-11088	27	24	transformations	transformation	NOUN
app01-11088	27	25	.	.	PUNCT
app01-11088	28	1	this	this	DET
app01-11088	28	2	idea	idea	NOUN
app01-11088	28	3	,	,	PUNCT
app01-11088	28	4	as	as	ADP
app01-11088	28	5	a	a	DET
app01-11088	28	6	consequence	consequence	NOUN
app01-11088	28	7	,	,	PUNCT
app01-11088	28	8	leads	lead	VERB
app01-11088	28	9	us	we	PRON
app01-11088	28	10	to	to	ADP
app01-11088	28	11	the	the	DET
app01-11088	28	12	possibility	possibility	NOUN
app01-11088	28	13	of	of	ADP
app01-11088	28	14	further	far	ADV
app01-11088	28	15	categorizing	categorize	VERB
app01-11088	28	16	geometric	geometric	ADJ
app01-11088	28	17	descriptors	descriptor	NOUN
app01-11088	28	18	17	17	NUM
app01-11088	28	19	https://doi.org/10.14311/app.2025.54.0017	https://doi.org/10.14311/app.2025.54.0017	NOUN
app01-11088	28	20	https://creativecommons.org/licenses/by/4.0/	https://creativecommons.org/licenses/by/4.0/	PROPN
app01-11088	29	1	https://www.cvut.cz/en	https://www.cvut.cz/en	PROPN
app01-11088	29	2	k.	k.	PROPN
app01-11088	30	1	das	das	PROPN
app01-11088	30	2	,	,	PUNCT
app01-11088	30	3	j.	j.	PROPN
app01-11088	30	4	sýkora	sýkora	PROPN
app01-11088	30	5	,	,	PUNCT
app01-11088	30	6	a.	a.	PROPN
app01-11088	30	7	kučerová	kučerová	PROPN
app01-11088	30	8	acta	acta	PROPN
app01-11088	30	9	polytechnica	polytechnica	PROPN
app01-11088	30	10	ctu	ctu	NOUN
app01-11088	30	11	proceedings	proceeding	NOUN
app01-11088	30	12	into	into	ADP
app01-11088	30	13	two	two	NUM
app01-11088	30	14	distinct	distinct	ADJ
app01-11088	30	15	classes	class	NOUN
app01-11088	30	16	:	:	PUNCT
app01-11088	30	17	first	first	ADV
app01-11088	30	18	,	,	PUNCT
app01-11088	30	19	descriptors	descriptor	NOUN
app01-11088	30	20	that	that	PRON
app01-11088	30	21	are	be	AUX
app01-11088	30	22	shape	shape	NOUN
app01-11088	30	23	specific	specific	ADJ
app01-11088	30	24	and	and	CCONJ
app01-11088	30	25	are	be	AUX
app01-11088	30	26	invariant	invariant	ADJ
app01-11088	30	27	to	to	ADP
app01-11088	30	28	certain	certain	ADJ
app01-11088	30	29	types	type	NOUN
app01-11088	30	30	of	of	ADP
app01-11088	30	31	transformations	transformation	NOUN
app01-11088	30	32	,	,	PUNCT
app01-11088	30	33	and	and	CCONJ
app01-11088	30	34	second	second	ADJ
app01-11088	30	35	,	,	PUNCT
app01-11088	30	36	descriptors	descriptor	NOUN
app01-11088	30	37	that	that	PRON
app01-11088	30	38	specify	specify	VERB
app01-11088	30	39	the	the	DET
app01-11088	30	40	state	state	NOUN
app01-11088	30	41	of	of	ADP
app01-11088	30	42	the	the	DET
app01-11088	30	43	shape	shape	NOUN
app01-11088	30	44	,	,	PUNCT
app01-11088	30	45	such	such	ADJ
app01-11088	30	46	as	as	ADP
app01-11088	30	47	size	size	NOUN
app01-11088	30	48	or	or	CCONJ
app01-11088	30	49	orientation	orientation	NOUN
app01-11088	30	50	etc	etc	X
app01-11088	30	51	.	.	X
app01-11088	31	1	hence	hence	ADV
app01-11088	31	2	,	,	PUNCT
app01-11088	31	3	from	from	ADP
app01-11088	31	4	hereon	hereon	NOUN
app01-11088	31	5	,	,	PUNCT
app01-11088	31	6	we	we	PRON
app01-11088	31	7	refer	refer	VERB
app01-11088	31	8	to	to	ADP
app01-11088	31	9	the	the	DET
app01-11088	31	10	first	first	ADJ
app01-11088	31	11	type	type	NOUN
app01-11088	31	12	as	as	ADP
app01-11088	31	13	shape	shape	NOUN
app01-11088	31	14	constant	constant	ADJ
app01-11088	31	15	(	(	PUNCT
app01-11088	31	16	sc	sc	PROPN
app01-11088	31	17	)	)	PUNCT
app01-11088	31	18	and	and	CCONJ
app01-11088	31	19	the	the	DET
app01-11088	31	20	second	second	ADJ
app01-11088	31	21	type	type	NOUN
app01-11088	31	22	as	as	ADP
app01-11088	31	23	state	state	NOUN
app01-11088	31	24	descriptors	descriptor	NOUN
app01-11088	31	25	(	(	PUNCT
app01-11088	31	26	sd	sd	NOUN
app01-11088	31	27	)	)	PUNCT
app01-11088	31	28	.	.	PUNCT
app01-11088	32	1	as	as	SCONJ
app01-11088	32	2	is	be	AUX
app01-11088	32	3	demonstrated	demonstrate	VERB
app01-11088	32	4	later	later	ADV
app01-11088	32	5	,	,	PUNCT
app01-11088	32	6	for	for	ADP
app01-11088	32	7	any	any	DET
app01-11088	32	8	shape	shape	NOUN
app01-11088	32	9	represented	represent	VERB
app01-11088	32	10	using	use	VERB
app01-11088	32	11	a	a	DET
app01-11088	32	12	set	set	NOUN
app01-11088	32	13	of	of	ADP
app01-11088	32	14	scs	scs	PROPN
app01-11088	32	15	,	,	PUNCT
app01-11088	32	16	sd	sd	NOUN
app01-11088	32	17	adds	add	VERB
app01-11088	32	18	a	a	DET
app01-11088	32	19	level	level	NOUN
app01-11088	32	20	of	of	ADP
app01-11088	32	21	tunability	tunability	NOUN
app01-11088	32	22	for	for	ADP
app01-11088	32	23	each	each	DET
app01-11088	32	24	shape	shape	NOUN
app01-11088	32	25	.	.	PUNCT
app01-11088	33	1	in	in	ADP
app01-11088	33	2	the	the	DET
app01-11088	33	3	subsequent	subsequent	ADJ
app01-11088	33	4	sections	section	NOUN
app01-11088	33	5	,	,	PUNCT
app01-11088	33	6	we	we	PRON
app01-11088	33	7	focus	focus	VERB
app01-11088	33	8	on	on	ADP
app01-11088	33	9	moments	moment	NOUN
app01-11088	33	10	of	of	ADP
app01-11088	33	11	binary	binary	ADJ
app01-11088	33	12	images	image	NOUN
app01-11088	33	13	that	that	PRON
app01-11088	33	14	are	be	AUX
app01-11088	33	15	invariant	invariant	ADJ
app01-11088	33	16	to	to	ADP
app01-11088	33	17	both	both	DET
app01-11088	33	18	translation	translation	NOUN
app01-11088	33	19	and	and	CCONJ
app01-11088	33	20	scale	scale	NOUN
app01-11088	33	21	,	,	PUNCT
app01-11088	33	22	and	and	CCONJ
app01-11088	33	23	treat	treat	VERB
app01-11088	33	24	them	they	PRON
app01-11088	33	25	as	as	ADP
app01-11088	33	26	shape	shape	NOUN
app01-11088	33	27	constants	constant	NOUN
app01-11088	33	28	.	.	PUNCT
app01-11088	34	1	subsequently	subsequently	ADV
app01-11088	34	2	,	,	PUNCT
app01-11088	34	3	we	we	PRON
app01-11088	34	4	train	train	VERB
app01-11088	34	5	a	a	DET
app01-11088	34	6	neural	neural	ADJ
app01-11088	34	7	network	network	NOUN
app01-11088	34	8	to	to	PART
app01-11088	34	9	generate	generate	VERB
app01-11088	34	10	shapes	shape	NOUN
app01-11088	34	11	corresponding	correspond	VERB
app01-11088	34	12	to	to	ADP
app01-11088	34	13	these	these	DET
app01-11088	34	14	moments	moment	NOUN
app01-11088	34	15	.	.	PUNCT
app01-11088	35	1	moreover	moreover	ADV
app01-11088	35	2	,	,	PUNCT
app01-11088	35	3	we	we	PRON
app01-11088	35	4	condition	condition	VERB
app01-11088	35	5	the	the	DET
app01-11088	35	6	network	network	NOUN
app01-11088	35	7	over	over	ADP
app01-11088	35	8	the	the	DET
app01-11088	35	9	area	area	NOUN
app01-11088	35	10	of	of	ADP
app01-11088	35	11	the	the	DET
app01-11088	35	12	shapes	shape	NOUN
app01-11088	35	13	such	such	ADJ
app01-11088	35	14	that	that	SCONJ
app01-11088	35	15	the	the	DET
app01-11088	35	16	network	network	NOUN
app01-11088	35	17	is	be	AUX
app01-11088	35	18	able	able	ADJ
app01-11088	35	19	to	to	PART
app01-11088	35	20	generate	generate	VERB
app01-11088	35	21	similar	similar	ADJ
app01-11088	35	22	shapes	shape	NOUN
app01-11088	35	23	of	of	ADP
app01-11088	35	24	varying	vary	VERB
app01-11088	35	25	sizes	size	NOUN
app01-11088	35	26	.	.	PUNCT
app01-11088	36	1	this	this	PRON
app01-11088	36	2	is	be	AUX
app01-11088	36	3	despite	despite	SCONJ
app01-11088	36	4	the	the	DET
app01-11088	36	5	fact	fact	NOUN
app01-11088	36	6	that	that	SCONJ
app01-11088	36	7	the	the	DET
app01-11088	36	8	data	data	NOUN
app01-11088	36	9	-	-	PUNCT
app01-11088	36	10	set	set	NOUN
app01-11088	36	11	is	be	AUX
app01-11088	36	12	not	not	PART
app01-11088	36	13	explicitly	explicitly	ADV
app01-11088	36	14	augmented	augment	VERB
app01-11088	36	15	to	to	PART
app01-11088	36	16	provide	provide	VERB
app01-11088	36	17	any	any	DET
app01-11088	36	18	relation	relation	NOUN
app01-11088	36	19	between	between	ADP
app01-11088	36	20	aggregate	aggregate	ADJ
app01-11088	36	21	size	size	NOUN
app01-11088	36	22	and	and	CCONJ
app01-11088	36	23	its	its	PRON
app01-11088	36	24	shape	shape	NOUN
app01-11088	36	25	constants	constant	NOUN
app01-11088	36	26	.	.	PUNCT
app01-11088	37	1	thereby	thereby	ADV
app01-11088	37	2	highlighting	highlight	VERB
app01-11088	37	3	the	the	DET
app01-11088	37	4	ability	ability	NOUN
app01-11088	37	5	of	of	ADP
app01-11088	37	6	the	the	DET
app01-11088	37	7	model	model	NOUN
app01-11088	37	8	to	to	PART
app01-11088	37	9	learn	learn	VERB
app01-11088	37	10	the	the	DET
app01-11088	37	11	mutual	mutual	ADJ
app01-11088	37	12	independence	independence	NOUN
app01-11088	37	13	between	between	ADP
app01-11088	37	14	the	the	DET
app01-11088	37	15	shape	shape	NOUN
app01-11088	37	16	invariants	invariant	NOUN
app01-11088	37	17	and	and	CCONJ
app01-11088	37	18	the	the	DET
app01-11088	37	19	chosen	choose	VERB
app01-11088	37	20	state	state	NOUN
app01-11088	37	21	descriptor	descriptor	NOUN
app01-11088	37	22	.	.	PUNCT
app01-11088	38	1	therefore	therefore	ADV
app01-11088	38	2	,	,	PUNCT
app01-11088	38	3	in	in	ADP
app01-11088	38	4	this	this	DET
app01-11088	38	5	work	work	NOUN
app01-11088	38	6	,	,	PUNCT
app01-11088	38	7	section	section	NOUN
app01-11088	38	8	2	2	NUM
app01-11088	38	9	presents	present	VERB
app01-11088	38	10	the	the	DET
app01-11088	38	11	overall	overall	ADJ
app01-11088	38	12	methodology	methodology	NOUN
app01-11088	38	13	of	of	ADP
app01-11088	38	14	obtining	obtine	VERB
app01-11088	38	15	the	the	DET
app01-11088	38	16	shape	shape	NOUN
app01-11088	38	17	invariants	invariant	NOUN
app01-11088	38	18	and	and	CCONJ
app01-11088	38	19	details	detail	NOUN
app01-11088	38	20	about	about	ADP
app01-11088	38	21	the	the	DET
app01-11088	38	22	neural	neural	ADJ
app01-11088	38	23	network	network	NOUN
app01-11088	38	24	architecture	architecture	NOUN
app01-11088	38	25	that	that	PRON
app01-11088	38	26	is	be	AUX
app01-11088	38	27	used	use	VERB
app01-11088	38	28	.	.	PUNCT
app01-11088	39	1	this	this	PRON
app01-11088	39	2	is	be	AUX
app01-11088	39	3	followed	follow	VERB
app01-11088	39	4	by	by	ADP
app01-11088	39	5	the	the	DET
app01-11088	39	6	results	result	NOUN
app01-11088	39	7	in	in	ADP
app01-11088	39	8	section	section	NOUN
app01-11088	39	9	3	3	NUM
app01-11088	39	10	and	and	CCONJ
app01-11088	39	11	is	be	AUX
app01-11088	39	12	finally	finally	ADV
app01-11088	39	13	followed	follow	VERB
app01-11088	39	14	by	by	ADP
app01-11088	39	15	the	the	DET
app01-11088	39	16	conclusion	conclusion	NOUN
app01-11088	39	17	derived	derive	VERB
app01-11088	39	18	from	from	ADP
app01-11088	39	19	this	this	DET
app01-11088	39	20	work	work	NOUN
app01-11088	39	21	in	in	ADP
app01-11088	39	22	section	section	NOUN
app01-11088	39	23	4	4	NUM
app01-11088	39	24	.	.	NOUN
app01-11088	39	25	2	2	NUM
app01-11088	39	26	.	.	X
app01-11088	39	27	methodology	methodology	NOUN
app01-11088	39	28	2.1	2.1	NUM
app01-11088	39	29	.	.	PUNCT
app01-11088	39	30	basic	basic	ADJ
app01-11088	39	31	theory	theory	NOUN
app01-11088	39	32	of	of	ADP
app01-11088	39	33	image	image	NOUN
app01-11088	39	34	moments	moment	NOUN
app01-11088	39	35	computation	computation	NOUN
app01-11088	39	36	of	of	ADP
app01-11088	39	37	image	image	NOUN
app01-11088	39	38	moments	moment	NOUN
app01-11088	39	39	relies	rely	VERB
app01-11088	39	40	on	on	ADP
app01-11088	39	41	defining	define	VERB
app01-11088	39	42	an	an	DET
app01-11088	39	43	image	image	NOUN
app01-11088	39	44	as	as	ADP
app01-11088	39	45	an	an	DET
app01-11088	39	46	intensity	intensity	NOUN
app01-11088	39	47	distribution	distribution	NOUN
app01-11088	39	48	function	function	NOUN
app01-11088	39	49	ρ(x	ρ(x	PROPN
app01-11088	39	50	,	,	PUNCT
app01-11088	39	51	y	y	NOUN
app01-11088	39	52	)	)	PUNCT
app01-11088	39	53	,	,	PUNCT
app01-11088	39	54	which	which	PRON
app01-11088	39	55	in	in	ADP
app01-11088	39	56	the	the	DET
app01-11088	39	57	case	case	NOUN
app01-11088	39	58	of	of	ADP
app01-11088	39	59	single	single	ADJ
app01-11088	39	60	-	-	PUNCT
app01-11088	39	61	channel	channel	NOUN
app01-11088	39	62	binary	binary	ADJ
app01-11088	39	63	images	image	NOUN
app01-11088	39	64	is	be	AUX
app01-11088	39	65	given	give	VERB
app01-11088	39	66	as	as	ADP
app01-11088	39	67	i(x	i(x	PROPN
app01-11088	39	68	,	,	PUNCT
app01-11088	39	69	y	y	PROPN
app01-11088	39	70	)	)	PUNCT
app01-11088	39	71	,	,	PUNCT
app01-11088	40	1	where	where	SCONJ
app01-11088	40	2	,	,	PUNCT
app01-11088	40	3	i(x	i(x	PROPN
app01-11088	40	4	,	,	PUNCT
app01-11088	40	5	y	y	PROPN
app01-11088	40	6	)	)	PUNCT
app01-11088	40	7	=	=	SYM
app01-11088	40	8	1	1	NUM
app01-11088	40	9	if	if	SCONJ
app01-11088	40	10	the	the	DET
app01-11088	40	11	point	point	NOUN
app01-11088	40	12	(	(	PUNCT
app01-11088	40	13	x	x	NOUN
app01-11088	40	14	,	,	PUNCT
app01-11088	40	15	y	y	NOUN
app01-11088	40	16	)	)	PUNCT
app01-11088	40	17	is	be	AUX
app01-11088	40	18	in	in	ADP
app01-11088	40	19	the	the	DET
app01-11088	40	20	region	region	NOUN
app01-11088	40	21	of	of	ADP
app01-11088	40	22	interest	interest	NOUN
app01-11088	40	23	or	or	CCONJ
app01-11088	40	24	else	else	ADV
app01-11088	40	25	i(x	i(x	PROPN
app01-11088	40	26	,	,	PUNCT
app01-11088	40	27	y	y	NOUN
app01-11088	40	28	)	)	PUNCT
app01-11088	41	1	=	=	NOUN
app01-11088	41	2	0	0	X
app01-11088	41	3	.	.	PUNCT
app01-11088	42	1	as	as	ADP
app01-11088	42	2	such	such	ADJ
app01-11088	42	3	,	,	PUNCT
app01-11088	42	4	the	the	DET
app01-11088	42	5	general	general	ADJ
app01-11088	42	6	expression	expression	NOUN
app01-11088	42	7	of	of	ADP
app01-11088	42	8	a	a	DET
app01-11088	42	9	generic	generic	ADJ
app01-11088	42	10	moment	moment	NOUN
app01-11088	42	11	mpq	mpq	X
app01-11088	42	12	of	of	ADP
app01-11088	42	13	(	(	PUNCT
app01-11088	42	14	p	p	X
app01-11088	43	1	+	+	CCONJ
app01-11088	44	1	q)th	q)th	ADJ
app01-11088	44	2	order	order	NOUN
app01-11088	44	3	is	be	AUX
app01-11088	44	4	given	give	VERB
app01-11088	44	5	as	as	ADP
app01-11088	44	6	:	:	PUNCT
app01-11088	44	7	mpq	mpq	X
app01-11088	44	8	=	=	SYM
app01-11088	44	9	n	n	CCONJ
app01-11088	44	10	p∑	p∑	NOUN
app01-11088	45	1	i=0	i=0	PROPN
app01-11088	45	2	q∑	q∑	PROPN
app01-11088	45	3	j=0	j=0	PROPN
app01-11088	45	4	k(x	k(x	PROPN
app01-11088	45	5	,	,	PUNCT
app01-11088	45	6	y)i(x	y)i(x	NOUN
app01-11088	45	7	,	,	PUNCT
app01-11088	45	8	y	y	PROPN
app01-11088	45	9	)	)	PUNCT
app01-11088	45	10	,	,	PUNCT
app01-11088	45	11	(	(	PUNCT
app01-11088	45	12	1	1	X
app01-11088	45	13	)	)	PUNCT
app01-11088	45	14	where	where	SCONJ
app01-11088	45	15	,	,	PUNCT
app01-11088	45	16	n	n	PRON
app01-11088	45	17	is	be	AUX
app01-11088	45	18	the	the	DET
app01-11088	45	19	normalizing	normalizing	ADJ
app01-11088	45	20	factor	factor	NOUN
app01-11088	45	21	and	and	CCONJ
app01-11088	45	22	k(x	k(x	PROPN
app01-11088	45	23	,	,	PUNCT
app01-11088	45	24	y	y	NOUN
app01-11088	45	25	)	)	PUNCT
app01-11088	45	26	is	be	AUX
app01-11088	45	27	the	the	DET
app01-11088	45	28	moment	moment	NOUN
app01-11088	45	29	kernel	kernel	PROPN
app01-11088	45	30	.	.	PUNCT
app01-11088	46	1	for	for	ADP
app01-11088	46	2	raw	raw	ADJ
app01-11088	46	3	-	-	PUNCT
app01-11088	46	4	geometric	geometric	ADJ
app01-11088	46	5	moments	moment	NOUN
app01-11088	46	6	,	,	PUNCT
app01-11088	46	7	denoted	denote	VERB
app01-11088	46	8	using	use	VERB
app01-11088	46	9	mpq	mpq	PROPN
app01-11088	46	10	,	,	PUNCT
app01-11088	46	11	n	n	NOUN
app01-11088	46	12	=	=	SYM
app01-11088	46	13	1	1	NUM
app01-11088	46	14	and	and	CCONJ
app01-11088	46	15	k(x	k(x	PROPN
app01-11088	46	16	,	,	PUNCT
app01-11088	46	17	y	y	NOUN
app01-11088	46	18	)	)	PUNCT
app01-11088	46	19	=	=	SYM
app01-11088	46	20	xpyq	xpyq	NOUN
app01-11088	46	21	.	.	PUNCT
app01-11088	47	1	the	the	DET
app01-11088	47	2	raw	raw	ADJ
app01-11088	47	3	moment	moment	NOUN
app01-11088	47	4	m00	m00	NOUN
app01-11088	47	5	indicates	indicate	VERB
app01-11088	47	6	the	the	DET
app01-11088	47	7	area	area	NOUN
app01-11088	47	8	of	of	ADP
app01-11088	47	9	the	the	DET
app01-11088	47	10	region	region	NOUN
app01-11088	47	11	of	of	ADP
app01-11088	47	12	interest	interest	NOUN
app01-11088	47	13	and	and	CCONJ
app01-11088	47	14	(	(	PUNCT
app01-11088	47	15	x̄	x̄	PROPN
app01-11088	47	16	,	,	PUNCT
app01-11088	47	17	ȳ	ȳ	PROPN
app01-11088	47	18	)	)	PUNCT
app01-11088	47	19	gives	give	VERB
app01-11088	47	20	its	its	PRON
app01-11088	47	21	centroid	centroid	NOUN
app01-11088	47	22	such	such	ADJ
app01-11088	47	23	that	that	PRON
app01-11088	47	24	x̄	x̄	NOUN
app01-11088	47	25	=	=	SYM
app01-11088	47	26	m10	m10	PROPN
app01-11088	47	27	/	/	SYM
app01-11088	47	28	m00	m00	NOUN
app01-11088	47	29	,	,	PUNCT
app01-11088	47	30	ȳ	ȳ	PROPN
app01-11088	47	31	=	=	SYM
app01-11088	47	32	m01	m01	PROPN
app01-11088	47	33	/	/	SYM
app01-11088	47	34	m00	m00	NOUN
app01-11088	47	35	.	.	PUNCT
app01-11088	48	1	since	since	SCONJ
app01-11088	48	2	this	this	DET
app01-11088	48	3	work	work	NOUN
app01-11088	48	4	particularly	particularly	ADV
app01-11088	48	5	focuses	focus	VERB
app01-11088	48	6	on	on	ADP
app01-11088	48	7	translation	translation	NOUN
app01-11088	48	8	and	and	CCONJ
app01-11088	48	9	scale	scale	NOUN
app01-11088	48	10	invariant	invariant	ADJ
app01-11088	48	11	moments	moment	NOUN
app01-11088	48	12	of	of	ADP
app01-11088	48	13	aggregates	aggregate	NOUN
app01-11088	48	14	,	,	PUNCT
app01-11088	48	15	for	for	ADP
app01-11088	48	16	any	any	DET
app01-11088	48	17	translation	translation	NOUN
app01-11088	48	18	invariant	invariant	ADJ
app01-11088	48	19	moment	moment	NOUN
app01-11088	48	20	,	,	PUNCT
app01-11088	48	21	also	also	ADV
app01-11088	48	22	known	know	VERB
app01-11088	48	23	as	as	ADP
app01-11088	48	24	central	central	ADJ
app01-11088	48	25	moment	moment	NOUN
app01-11088	48	26	,	,	PUNCT
app01-11088	48	27	and	and	CCONJ
app01-11088	48	28	denoted	denote	VERB
app01-11088	48	29	using	use	VERB
app01-11088	48	30	µpq	µpq	ADJ
app01-11088	48	31	,	,	PUNCT
app01-11088	48	32	in	in	ADP
app01-11088	48	33	which	which	DET
app01-11088	48	34	case	case	NOUN
app01-11088	48	35	,	,	PUNCT
app01-11088	48	36	n	n	NOUN
app01-11088	48	37	=	=	SYM
app01-11088	48	38	1	1	NUM
app01-11088	48	39	and	and	CCONJ
app01-11088	48	40	k(x	k(x	PROPN
app01-11088	48	41	,	,	PUNCT
app01-11088	48	42	y	y	NOUN
app01-11088	48	43	)	)	PUNCT
app01-11088	48	44	=	=	SYM
app01-11088	49	1	(	(	PUNCT
app01-11088	49	2	x	x	X
app01-11088	49	3	−	−	PROPN
app01-11088	49	4	x̄)p(y	x̄)p(y	INTJ
app01-11088	49	5	−	−	PROPN
app01-11088	49	6	ȳ)q	ȳ)q	NOUN
app01-11088	49	7	,	,	PUNCT
app01-11088	49	8	the	the	DET
app01-11088	49	9	scale	scale	NOUN
app01-11088	49	10	and	and	CCONJ
app01-11088	49	11	translation	translation	NOUN
app01-11088	49	12	invariant	invariant	ADJ
app01-11088	49	13	moment	moment	NOUN
app01-11088	49	14	,	,	PUNCT
app01-11088	49	15	also	also	ADV
app01-11088	49	16	known	know	VERB
app01-11088	49	17	as	as	ADP
app01-11088	49	18	normalizedcentral	normalizedcentral	ADJ
app01-11088	49	19	moments	moment	NOUN
app01-11088	49	20	are	be	AUX
app01-11088	49	21	given	give	VERB
app01-11088	49	22	by	by	ADP
app01-11088	49	23	:	:	PUNCT
app01-11088	49	24	ηpq	ηpq	PROPN
app01-11088	49	25	=	=	SYM
app01-11088	49	26	1	1	NUM
app01-11088	49	27	µ	µ	X
app01-11088	49	28	(	(	PUNCT
app01-11088	49	29	1	1	NUM
app01-11088	49	30	+	+	NUM
app01-11088	49	31	p+q	p+q	NUM
app01-11088	49	32	2	2	NUM
app01-11088	49	33	)	)	PUNCT
app01-11088	49	34	00	00	PUNCT
app01-11088	49	35	µpq	µpq	ADJ
app01-11088	49	36	.	.	PUNCT
app01-11088	50	1	(	(	PUNCT
app01-11088	50	2	2	2	X
app01-11088	50	3	)	)	PUNCT
app01-11088	50	4	s	s	VERB
app01-11088	50	5	h	h	NOUN
app01-11088	50	6	a	a	DET
app01-11088	50	7	p	p	X
app01-11088	50	8	e	e	NOUN
app01-11088	50	9	c	c	NOUN
app01-11088	50	10	o	o	PROPN
app01-11088	51	1	n	n	X
app01-11088	51	2	s	s	PROPN
app01-11088	51	3	t	t	NOUN
app01-11088	51	4	a	a	PRON
app01-11088	51	5	n	n	NOUN
app01-11088	51	6	t	t	NOUN
app01-11088	51	7	s	s	NOUN
app01-11088	51	8	s	s	PROPN
app01-11088	51	9	t	t	PROPN
app01-11088	52	1	a	a	DET
app01-11088	52	2	t	t	NOUN
app01-11088	52	3	e	e	X
app01-11088	52	4	d	d	X
app01-11088	52	5	e	e	X
app01-11088	52	6	s	s	PROPN
app01-11088	52	7	c	c	NOUN
app01-11088	52	8	r	r	NOUN
app01-11088	53	1	i	i	PRON
app01-11088	53	2	p	p	NOUN
app01-11088	54	1	t	t	X
app01-11088	54	2	o	o	X
app01-11088	54	3	r	r	NOUN
app01-11088	54	4	s	s	PROPN
app01-11088	54	5	encoder	encoder	NOUN
app01-11088	54	6	head	head	NOUN
app01-11088	54	7	1	1	NUM
app01-11088	54	8	encoder	encoder	NOUN
app01-11088	54	9	head	head	NOUN
app01-11088	54	10	2	2	NUM
app01-11088	54	11	decoder	decoder	NOUN
app01-11088	54	12	encoder	encoder	NOUN
app01-11088	54	13	outputs	output	NOUN
app01-11088	54	14	stack	stack	VERB
app01-11088	54	15	outputdecoder	outputdecoder	NOUN
app01-11088	54	16	input	input	NOUN
app01-11088	54	17	figure	figure	NOUN
app01-11088	54	18	1	1	NUM
app01-11088	54	19	.	.	PUNCT
app01-11088	54	20	general	general	ADJ
app01-11088	54	21	architecture	architecture	NOUN
app01-11088	54	22	of	of	ADP
app01-11088	54	23	the	the	DET
app01-11088	54	24	aggregate	aggregate	ADJ
app01-11088	54	25	generation	generation	NOUN
app01-11088	54	26	network	network	NOUN
app01-11088	54	27	.	.	PUNCT
app01-11088	55	1	consequently	consequently	ADV
app01-11088	55	2	,	,	PUNCT
app01-11088	55	3	the	the	DET
app01-11088	55	4	final	final	ADJ
app01-11088	55	5	expression	expression	NOUN
app01-11088	55	6	for	for	ADP
app01-11088	55	7	scale	scale	NOUN
app01-11088	55	8	and	and	CCONJ
app01-11088	55	9	translation	translation	NOUN
app01-11088	55	10	invariant	invariant	ADJ
app01-11088	55	11	moments	moment	NOUN
app01-11088	55	12	i.e.	i.e.	X
app01-11088	55	13	normalized	normalize	VERB
app01-11088	55	14	-	-	PUNCT
app01-11088	55	15	central	central	ADJ
app01-11088	55	16	moments	moment	NOUN
app01-11088	55	17	is	be	AUX
app01-11088	55	18	given	give	VERB
app01-11088	55	19	as	as	ADP
app01-11088	55	20	:	:	PUNCT
app01-11088	55	21	ηpq	ηpq	PROPN
app01-11088	55	22	=	=	SYM
app01-11088	55	23	1	1	NUM
app01-11088	55	24	µ	µ	X
app01-11088	55	25	(	(	PUNCT
app01-11088	55	26	1	1	NUM
app01-11088	55	27	+	+	NUM
app01-11088	55	28	p+q	p+q	NUM
app01-11088	55	29	2	2	NUM
app01-11088	55	30	)	)	PUNCT
app01-11088	55	31	00	00	PUNCT
app01-11088	56	1	p∑	p∑	INTJ
app01-11088	57	1	i=0	i=0	PROPN
app01-11088	57	2	q∑	q∑	PROPN
app01-11088	57	3	j=0	j=0	PROPN
app01-11088	57	4	(	(	PUNCT
app01-11088	57	5	x	x	X
app01-11088	57	6	−	−	PROPN
app01-11088	58	1	x̄)p(y	x̄)p(y	INTJ
app01-11088	58	2	−	−	PROPN
app01-11088	59	1	ȳ)qi(x	ȳ)qi(x	PROPN
app01-11088	59	2	,	,	PUNCT
app01-11088	59	3	y	y	PROPN
app01-11088	59	4	)	)	PUNCT
app01-11088	59	5	.	.	PUNCT
app01-11088	60	1	(	(	PUNCT
app01-11088	60	2	3	3	X
app01-11088	60	3	)	)	PUNCT
app01-11088	60	4	in	in	ADP
app01-11088	60	5	the	the	DET
app01-11088	60	6	subsequent	subsequent	ADJ
app01-11088	60	7	work	work	NOUN
app01-11088	60	8	,	,	PUNCT
app01-11088	60	9	moments	moment	NOUN
app01-11088	60	10	derived	derive	VERB
app01-11088	60	11	using	use	VERB
app01-11088	60	12	equation	equation	NOUN
app01-11088	60	13	3	3	NUM
app01-11088	60	14	is	be	AUX
app01-11088	60	15	used	use	VERB
app01-11088	60	16	for	for	ADP
app01-11088	60	17	quantification	quantification	NOUN
app01-11088	60	18	of	of	ADP
app01-11088	60	19	aggregate	aggregate	ADJ
app01-11088	60	20	shapes	shape	NOUN
app01-11088	60	21	,	,	PUNCT
app01-11088	60	22	while	while	SCONJ
app01-11088	60	23	m00	m00	NOUN
app01-11088	60	24	is	be	AUX
app01-11088	60	25	used	use	VERB
app01-11088	60	26	as	as	ADP
app01-11088	60	27	the	the	DET
app01-11088	60	28	state	state	NOUN
app01-11088	60	29	descriptor	descriptor	NOUN
app01-11088	60	30	for	for	ADP
app01-11088	60	31	the	the	DET
app01-11088	60	32	area	area	NOUN
app01-11088	60	33	.	.	PUNCT
app01-11088	61	1	it	it	PRON
app01-11088	61	2	is	be	AUX
app01-11088	61	3	to	to	PART
app01-11088	61	4	be	be	AUX
app01-11088	61	5	noted	note	VERB
app01-11088	61	6	that	that	SCONJ
app01-11088	61	7	η00	η00	PROPN
app01-11088	61	8	=	=	SYM
app01-11088	61	9	1	1	NUM
app01-11088	61	10	and	and	CCONJ
app01-11088	61	11	η01	η01	NOUN
app01-11088	61	12	=	=	SYM
app01-11088	61	13	η10	η10	NOUN
app01-11088	61	14	=	=	SYM
app01-11088	61	15	0	0	PUNCT
app01-11088	61	16	and	and	CCONJ
app01-11088	61	17	are	be	AUX
app01-11088	61	18	therefore	therefore	ADV
app01-11088	61	19	not	not	PART
app01-11088	61	20	used	use	VERB
app01-11088	61	21	.	.	PUNCT
app01-11088	62	1	therefore	therefore	ADV
app01-11088	62	2	,	,	PUNCT
app01-11088	62	3	the	the	DET
app01-11088	62	4	sets	set	NOUN
app01-11088	62	5	of	of	ADP
app01-11088	62	6	image	image	NOUN
app01-11088	62	7	moments	moment	NOUN
app01-11088	62	8	used	use	VERB
app01-11088	62	9	are	be	AUX
app01-11088	62	10	given	give	VERB
app01-11088	62	11	as	as	ADP
app01-11088	62	12	{	{	PUNCT
app01-11088	62	13	m00	m00	NOUN
app01-11088	62	14	}	}	PUNCT
app01-11088	62	15	,	,	PUNCT
app01-11088	62	16	which	which	PRON
app01-11088	62	17	is	be	AUX
app01-11088	62	18	the	the	DET
app01-11088	62	19	state	state	NOUN
app01-11088	62	20	descriptor	descriptor	NOUN
app01-11088	62	21	(	(	PUNCT
app01-11088	62	22	sd	sd	NOUN
app01-11088	62	23	)	)	PUNCT
app01-11088	62	24	and	and	CCONJ
app01-11088	62	25	{	{	PUNCT
app01-11088	62	26	η20	η20	NOUN
app01-11088	62	27	,	,	PUNCT
app01-11088	62	28	η11	η11	NOUN
app01-11088	62	29	,	,	PUNCT
app01-11088	62	30	·	·	PUNCT
app01-11088	62	31	·	·	PUNCT
app01-11088	62	32	·	·	PUNCT
app01-11088	62	33	,	,	PUNCT
app01-11088	62	34	η70	η70	NOUN
app01-11088	62	35	,	,	PUNCT
app01-11088	62	36	η61	η61	NOUN
app01-11088	62	37	,	,	PUNCT
app01-11088	62	38	η52	η52	NOUN
app01-11088	62	39	,	,	PUNCT
app01-11088	62	40	·	·	PUNCT
app01-11088	62	41	·	·	PUNCT
app01-11088	62	42	·	·	PUNCT
app01-11088	62	43	,	,	PUNCT
app01-11088	62	44	η07	η07	PROPN
app01-11088	62	45	}	}	PUNCT
app01-11088	62	46	,	,	PUNCT
app01-11088	62	47	which	which	PRON
app01-11088	62	48	is	be	AUX
app01-11088	62	49	the	the	DET
app01-11088	62	50	set	set	NOUN
app01-11088	62	51	of	of	ADP
app01-11088	62	52	shape	shape	NOUN
app01-11088	62	53	constants	constant	NOUN
app01-11088	62	54	(	(	PUNCT
app01-11088	62	55	scs	scs	PROPN
app01-11088	62	56	)	)	PUNCT
app01-11088	62	57	.	.	PUNCT
app01-11088	63	1	2.2	2.2	NUM
app01-11088	63	2	.	.	PUNCT
app01-11088	63	3	neural	neural	ADJ
app01-11088	63	4	network	network	NOUN
app01-11088	63	5	for	for	ADP
app01-11088	63	6	aggregate	aggregate	ADJ
app01-11088	63	7	generation	generation	NOUN
app01-11088	63	8	given	give	VERB
app01-11088	63	9	a	a	DET
app01-11088	63	10	set	set	NOUN
app01-11088	63	11	of	of	ADP
app01-11088	63	12	sd	sd	NOUN
app01-11088	63	13	and	and	CCONJ
app01-11088	63	14	sc	sc	PROPN
app01-11088	63	15	,	,	PUNCT
app01-11088	63	16	the	the	DET
app01-11088	63	17	objective	objective	NOUN
app01-11088	63	18	of	of	ADP
app01-11088	63	19	this	this	DET
app01-11088	63	20	work	work	NOUN
app01-11088	63	21	is	be	AUX
app01-11088	63	22	to	to	PART
app01-11088	63	23	demonstrate	demonstrate	VERB
app01-11088	63	24	the	the	DET
app01-11088	63	25	ability	ability	NOUN
app01-11088	63	26	of	of	ADP
app01-11088	63	27	a	a	DET
app01-11088	63	28	deep	deep	ADJ
app01-11088	63	29	neural	neural	ADJ
app01-11088	63	30	network	network	NOUN
app01-11088	63	31	to	to	PART
app01-11088	63	32	learn	learn	VERB
app01-11088	63	33	and	and	CCONJ
app01-11088	63	34	generate	generate	VERB
app01-11088	63	35	the	the	DET
app01-11088	63	36	shape	shape	NOUN
app01-11088	63	37	that	that	PRON
app01-11088	63	38	best	well	ADV
app01-11088	63	39	fits	fit	VERB
app01-11088	63	40	the	the	DET
app01-11088	63	41	description	description	NOUN
app01-11088	63	42	given	give	VERB
app01-11088	63	43	using	using	NOUN
app01-11088	63	44	and	and	CCONJ
app01-11088	63	45	sd	sd	NOUN
app01-11088	63	46	and	and	CCONJ
app01-11088	63	47	sc	sc	PROPN
app01-11088	63	48	.	.	PUNCT
app01-11088	64	1	to	to	ADP
app01-11088	64	2	this	this	DET
app01-11088	64	3	effect	effect	NOUN
app01-11088	64	4	,	,	PUNCT
app01-11088	64	5	a	a	DET
app01-11088	64	6	multi	multi	ADJ
app01-11088	64	7	-	-	ADJ
app01-11088	64	8	headed	headed	ADJ
app01-11088	64	9	autoencoder	autoencoder	NOUN
app01-11088	64	10	is	be	AUX
app01-11088	64	11	implemented	implement	VERB
app01-11088	64	12	that	that	PRON
app01-11088	64	13	makes	make	VERB
app01-11088	64	14	use	use	NOUN
app01-11088	64	15	of	of	ADP
app01-11088	64	16	two	two	NUM
app01-11088	64	17	encoder	encoder	NOUN
app01-11088	64	18	-	-	PUNCT
app01-11088	64	19	heads	head	NOUN
app01-11088	64	20	to	to	PART
app01-11088	64	21	encode	encode	VERB
app01-11088	64	22	the	the	DET
app01-11088	64	23	sc	sc	NOUN
app01-11088	64	24	and	and	CCONJ
app01-11088	64	25	sd	sd	NOUN
app01-11088	64	26	respectively	respectively	ADV
app01-11088	64	27	.	.	PUNCT
app01-11088	65	1	the	the	DET
app01-11088	65	2	outputs	output	NOUN
app01-11088	65	3	of	of	ADP
app01-11088	65	4	each	each	DET
app01-11088	65	5	encoder	encoder	NOUN
app01-11088	65	6	head	head	NOUN
app01-11088	65	7	are	be	AUX
app01-11088	65	8	combined	combine	VERB
app01-11088	65	9	and	and	CCONJ
app01-11088	65	10	used	use	VERB
app01-11088	65	11	as	as	ADP
app01-11088	65	12	input	input	NOUN
app01-11088	65	13	for	for	ADP
app01-11088	65	14	a	a	DET
app01-11088	65	15	convolution	convolution	NOUN
app01-11088	65	16	-	-	PUNCT
app01-11088	65	17	based	base	VERB
app01-11088	65	18	decoder	decoder	NOUN
app01-11088	65	19	section	section	NOUN
app01-11088	65	20	.	.	PUNCT
app01-11088	66	1	the	the	DET
app01-11088	66	2	overall	overall	ADJ
app01-11088	66	3	architecture	architecture	NOUN
app01-11088	66	4	of	of	ADP
app01-11088	66	5	the	the	DET
app01-11088	66	6	model	model	NOUN
app01-11088	66	7	is	be	AUX
app01-11088	66	8	presented	present	VERB
app01-11088	66	9	in	in	ADP
app01-11088	66	10	figure	figure	NOUN
app01-11088	66	11	1	1	NUM
app01-11088	66	12	.	.	PUNCT
app01-11088	67	1	each	each	PRON
app01-11088	67	2	of	of	ADP
app01-11088	67	3	the	the	DET
app01-11088	67	4	encoder	encoder	NOUN
app01-11088	67	5	heads	head	NOUN
app01-11088	67	6	is	be	AUX
app01-11088	67	7	composed	compose	VERB
app01-11088	67	8	of	of	ADP
app01-11088	67	9	a	a	DET
app01-11088	67	10	series	series	NOUN
app01-11088	67	11	of	of	ADP
app01-11088	67	12	fully	fully	ADV
app01-11088	67	13	-	-	PUNCT
app01-11088	67	14	connceted	conncete	VERB
app01-11088	67	15	layers	layer	NOUN
app01-11088	67	16	,	,	PUNCT
app01-11088	67	17	each	each	PRON
app01-11088	67	18	of	of	ADP
app01-11088	67	19	which	which	PRON
app01-11088	67	20	apply	apply	VERB
app01-11088	67	21	an	an	DET
app01-11088	67	22	affine	affine	NOUN
app01-11088	67	23	transform	transform	NOUN
app01-11088	67	24	to	to	ADP
app01-11088	67	25	the	the	DET
app01-11088	67	26	incoming	incoming	ADJ
app01-11088	67	27	vector	vector	NOUN
app01-11088	67	28	data	datum	NOUN
app01-11088	67	29	.	.	PUNCT
app01-11088	68	1	for	for	ADP
app01-11088	68	2	any	any	DET
app01-11088	68	3	incoming	incoming	ADJ
app01-11088	68	4	vector	vector	NOUN
app01-11088	68	5	u	u	NOUN
app01-11088	68	6	,	,	PUNCT
app01-11088	68	7	such	such	DET
app01-11088	68	8	an	an	DET
app01-11088	68	9	affine	affine	NOUN
app01-11088	68	10	transform	transform	NOUN
app01-11088	68	11	is	be	AUX
app01-11088	68	12	given	give	VERB
app01-11088	68	13	by	by	ADP
app01-11088	68	14	v	v	NOUN
app01-11088	68	15	=	=	SYM
app01-11088	68	16	uat	uat	NOUN
app01-11088	68	17	+	+	CCONJ
app01-11088	68	18	b	b	NOUN
app01-11088	68	19	,	,	PUNCT
app01-11088	68	20	where	where	SCONJ
app01-11088	68	21	v	v	NOUN
app01-11088	68	22	is	be	AUX
app01-11088	68	23	the	the	DET
app01-11088	68	24	output	output	NOUN
app01-11088	68	25	vector	vector	NOUN
app01-11088	68	26	and	and	CCONJ
app01-11088	68	27	b	b	NOUN
app01-11088	68	28	is	be	AUX
app01-11088	68	29	a	a	DET
app01-11088	68	30	vector	vector	NOUN
app01-11088	68	31	of	of	ADP
app01-11088	68	32	constants	constant	NOUN
app01-11088	68	33	,	,	PUNCT
app01-11088	68	34	also	also	ADV
app01-11088	68	35	known	know	VERB
app01-11088	68	36	as	as	ADP
app01-11088	68	37	bias	bias	NOUN
app01-11088	68	38	.	.	PUNCT
app01-11088	69	1	each	each	PRON
app01-11088	69	2	of	of	ADP
app01-11088	69	3	the	the	DET
app01-11088	69	4	fully	fully	ADV
app01-11088	69	5	-	-	PUNCT
app01-11088	69	6	connected	connect	VERB
app01-11088	69	7	layers	layer	NOUN
app01-11088	69	8	is	be	AUX
app01-11088	69	9	followed	follow	VERB
app01-11088	69	10	by	by	ADP
app01-11088	69	11	a	a	DET
app01-11088	69	12	rectified	rectified	ADJ
app01-11088	69	13	linear	linear	NOUN
app01-11088	69	14	unit	unit	NOUN
app01-11088	69	15	(	(	PUNCT
app01-11088	69	16	relu	relu	NOUN
app01-11088	69	17	)	)	PUNCT
app01-11088	69	18	,	,	PUNCT
app01-11088	69	19	which	which	PRON
app01-11088	69	20	imparts	impart	VERB
app01-11088	69	21	non	non	NOUN
app01-11088	69	22	-	-	ADJ
app01-11088	69	23	linearity	linearity	ADJ
app01-11088	69	24	to	to	ADP
app01-11088	69	25	the	the	DET
app01-11088	69	26	model	model	NOUN
app01-11088	69	27	.	.	PUNCT
app01-11088	70	1	relu	relu	NOUN
app01-11088	70	2	is	be	AUX
app01-11088	70	3	defined	define	VERB
app01-11088	70	4	as	as	ADP
app01-11088	70	5	the	the	DET
app01-11088	70	6	non	non	ADJ
app01-11088	70	7	-	-	ADJ
app01-11088	70	8	negative	negative	ADJ
app01-11088	70	9	part	part	NOUN
app01-11088	70	10	of	of	ADP
app01-11088	70	11	any	any	DET
app01-11088	70	12	argument	argument	NOUN
app01-11088	70	13	and	and	CCONJ
app01-11088	70	14	as	as	ADP
app01-11088	70	15	such	such	ADJ
app01-11088	70	16	,	,	PUNCT
app01-11088	70	17	relu(x	relu(x	PROPN
app01-11088	70	18	)	)	PUNCT
app01-11088	71	1	=	=	PUNCT
app01-11088	71	2	x	x	SYM
app01-11088	71	3	if	if	SCONJ
app01-11088	71	4	x	x	X
app01-11088	71	5	≥	≥	X
app01-11088	71	6	0	0	NUM
app01-11088	71	7	and	and	CCONJ
app01-11088	71	8	0	0	NUM
app01-11088	71	9	otherwise	otherwise	ADV
app01-11088	71	10	.	.	PUNCT
app01-11088	72	1	the	the	DET
app01-11088	72	2	transformation	transformation	NOUN
app01-11088	72	3	matrix	matrix	NOUN
app01-11088	72	4	a	a	PRON
app01-11088	72	5	and	and	CCONJ
app01-11088	72	6	the	the	DET
app01-11088	72	7	bias	bias	NOUN
app01-11088	72	8	vector	vector	PROPN
app01-11088	72	9	b	b	PROPN
app01-11088	72	10	is	be	AUX
app01-11088	72	11	optimized	optimize	VERB
app01-11088	72	12	through	through	ADP
app01-11088	72	13	training	training	NOUN
app01-11088	72	14	.	.	PUNCT
app01-11088	73	1	details	detail	NOUN
app01-11088	73	2	about	about	ADP
app01-11088	73	3	the	the	DET
app01-11088	73	4	encoder	encoder	NOUN
app01-11088	73	5	head	head	NOUN
app01-11088	73	6	1	1	NUM
app01-11088	73	7	is	be	AUX
app01-11088	73	8	given	give	VERB
app01-11088	73	9	in	in	ADP
app01-11088	73	10	table	table	NOUN
app01-11088	73	11	1	1	NUM
app01-11088	73	12	and	and	CCONJ
app01-11088	73	13	encoder	encoder	NOUN
app01-11088	73	14	head	head	NOUN
app01-11088	73	15	2	2	NUM
app01-11088	73	16	is	be	AUX
app01-11088	73	17	given	give	VERB
app01-11088	73	18	in	in	ADP
app01-11088	73	19	table	table	NOUN
app01-11088	73	20	2	2	NUM
app01-11088	73	21	.	.	PUNCT
app01-11088	74	1	both	both	CCONJ
app01-11088	74	2	the	the	DET
app01-11088	74	3	encoders	encoder	NOUN
app01-11088	74	4	are	be	AUX
app01-11088	74	5	designed	design	VERB
app01-11088	74	6	to	to	PART
app01-11088	74	7	have	have	VERB
app01-11088	74	8	output	output	NOUN
app01-11088	74	9	vectors	vector	NOUN
app01-11088	74	10	of	of	ADP
app01-11088	74	11	18	18	NUM
app01-11088	74	12	vol	vol	NOUN
app01-11088	74	13	.	.	PUNCT
app01-11088	75	1	54/2025	54/2025	NUM
app01-11088	75	2	tractable	tractable	ADJ
app01-11088	75	3	descriptors	descriptor	NOUN
app01-11088	75	4	for	for	ADP
app01-11088	75	5	digital	digital	ADJ
app01-11088	75	6	aggregate	aggregate	ADJ
app01-11088	75	7	generation	generation	NOUN
app01-11088	75	8	layer	layer	NOUN
app01-11088	75	9	in	in	ADP
app01-11088	75	10	/	/	SYM
app01-11088	75	11	out	out	ADP
app01-11088	75	12	-	-	PUNCT
app01-11088	75	13	features	feature	NOUN
app01-11088	75	14	activation	activation	NOUN
app01-11088	75	15	func	func	NOUN
app01-11088	75	16	.	.	PUNCT
app01-11088	76	1	fully	fully	ADV
app01-11088	76	2	connected	connect	VERB
app01-11088	76	3	1/1	1/1	NUM
app01-11088	76	4	024	024	NUM
app01-11088	76	5	relu	relu	NOUN
app01-11088	76	6	fully	fully	ADV
app01-11088	76	7	connected	connect	VERB
app01-11088	76	8	1	1	NUM
app01-11088	76	9	024/1	024/1	NUM
app01-11088	76	10	024	024	NUM
app01-11088	76	11	relu	relu	NOUN
app01-11088	76	12	table	table	NOUN
app01-11088	76	13	1	1	NUM
app01-11088	76	14	.	.	PUNCT
app01-11088	76	15	architecture	architecture	NOUN
app01-11088	76	16	of	of	ADP
app01-11088	76	17	encoder	encoder	NOUN
app01-11088	76	18	head	head	VERB
app01-11088	76	19	1	1	NUM
app01-11088	76	20	.	.	PUNCT
app01-11088	77	1	layer	layer	NOUN
app01-11088	77	2	in	in	ADP
app01-11088	77	3	/	/	SYM
app01-11088	77	4	out	out	ADP
app01-11088	77	5	-	-	PUNCT
app01-11088	77	6	features	feature	NOUN
app01-11088	77	7	activation	activation	NOUN
app01-11088	77	8	func	func	NOUN
app01-11088	77	9	.	.	PUNCT
app01-11088	78	1	fully	fully	ADV
app01-11088	78	2	connected	connect	VERB
app01-11088	78	3	33/2	33/2	NUM
app01-11088	78	4	048	048	NUM
app01-11088	78	5	relu	relu	NOUN
app01-11088	78	6	fully	fully	ADV
app01-11088	78	7	connected	connect	VERB
app01-11088	78	8	2	2	NUM
app01-11088	78	9	048/2	048/2	SYM
app01-11088	78	10	048	048	NUM
app01-11088	78	11	relu	relu	NOUN
app01-11088	78	12	fully	fully	ADV
app01-11088	78	13	connected	connect	VERB
app01-11088	78	14	2	2	NUM
app01-11088	78	15	048/1	048/1	NOUN
app01-11088	78	16	024	024	NUM
app01-11088	78	17	relu	relu	NOUN
app01-11088	78	18	fully	fully	ADV
app01-11088	78	19	connected	connect	VERB
app01-11088	78	20	1	1	NUM
app01-11088	78	21	024/1	024/1	NUM
app01-11088	78	22	024	024	NUM
app01-11088	78	23	relu	relu	NOUN
app01-11088	78	24	table	table	NOUN
app01-11088	78	25	2	2	NUM
app01-11088	78	26	.	.	PUNCT
app01-11088	78	27	architecture	architecture	NOUN
app01-11088	78	28	of	of	ADP
app01-11088	78	29	encoder	encoder	NOUN
app01-11088	78	30	head	head	NOUN
app01-11088	78	31	2	2	NUM
app01-11088	78	32	.	.	PUNCT
app01-11088	78	33	equal	equal	ADJ
app01-11088	78	34	size	size	NOUN
app01-11088	78	35	.	.	PUNCT
app01-11088	79	1	each	each	PRON
app01-11088	79	2	of	of	ADP
app01-11088	79	3	the	the	DET
app01-11088	79	4	output	output	NOUN
app01-11088	79	5	vectors	vector	NOUN
app01-11088	79	6	are	be	AUX
app01-11088	79	7	re	re	VERB
app01-11088	79	8	-	-	VERB
app01-11088	79	9	shaped	shape	VERB
app01-11088	79	10	into	into	ADP
app01-11088	79	11	a	a	DET
app01-11088	79	12	matrix	matrix	NOUN
app01-11088	79	13	and	and	CCONJ
app01-11088	79	14	concatenated	concatenate	VERB
app01-11088	79	15	to	to	PART
app01-11088	79	16	obtain	obtain	VERB
app01-11088	79	17	a	a	DET
app01-11088	79	18	tensor	tensor	NOUN
app01-11088	79	19	of	of	ADP
app01-11088	79	20	shape	shape	NOUN
app01-11088	79	21	[	[	X
app01-11088	79	22	2	2	NUM
app01-11088	79	23	,	,	PUNCT
app01-11088	79	24	32	32	NUM
app01-11088	79	25	,	,	PUNCT
app01-11088	79	26	32	32	NUM
app01-11088	79	27	]	]	PUNCT
app01-11088	79	28	,	,	PUNCT
app01-11088	79	29	which	which	PRON
app01-11088	79	30	forms	form	VERB
app01-11088	79	31	the	the	DET
app01-11088	79	32	input	input	NOUN
app01-11088	79	33	for	for	ADP
app01-11088	79	34	the	the	DET
app01-11088	79	35	decoder	decoder	NOUN
app01-11088	79	36	.	.	PUNCT
app01-11088	80	1	the	the	DET
app01-11088	80	2	first	first	ADJ
app01-11088	80	3	dimension	dimension	NOUN
app01-11088	80	4	here	here	ADV
app01-11088	80	5	is	be	AUX
app01-11088	80	6	termed	term	VERB
app01-11088	80	7	as	as	SCONJ
app01-11088	80	8	the	the	DET
app01-11088	80	9	feature	feature	NOUN
app01-11088	80	10	dimension	dimension	NOUN
app01-11088	80	11	and	and	CCONJ
app01-11088	80	12	the	the	DET
app01-11088	80	13	remaining	remain	VERB
app01-11088	80	14	dimensions	dimension	NOUN
app01-11088	80	15	are	be	AUX
app01-11088	80	16	the	the	DET
app01-11088	80	17	y	y	PROPN
app01-11088	80	18	and	and	CCONJ
app01-11088	80	19	x	x	PROPN
app01-11088	80	20	directions	direction	NOUN
app01-11088	80	21	in	in	ADP
app01-11088	80	22	the	the	DET
app01-11088	80	23	cartesian	cartesian	ADJ
app01-11088	80	24	coordinate	coordinate	NOUN
app01-11088	80	25	system	system	NOUN
app01-11088	80	26	respectively	respectively	ADV
app01-11088	80	27	.	.	PUNCT
app01-11088	81	1	the	the	DET
app01-11088	81	2	decoder	decoder	NOUN
app01-11088	81	3	comprises	comprise	NOUN
app01-11088	81	4	of	of	ADP
app01-11088	81	5	alternating	alternate	VERB
app01-11088	81	6	transposedconvolutions	transposedconvolution	NOUN
app01-11088	81	7	and	and	CCONJ
app01-11088	81	8	convolution	convolution	NOUN
app01-11088	81	9	layers	layer	NOUN
app01-11088	81	10	[	[	X
app01-11088	81	11	17	17	NUM
app01-11088	81	12	]	]	PUNCT
app01-11088	81	13	.	.	PUNCT
app01-11088	82	1	transposed	transpose	VERB
app01-11088	82	2	2d	2d	NOUN
app01-11088	82	3	-	-	PUNCT
app01-11088	82	4	convolution	convolution	NOUN
app01-11088	82	5	layers	layer	NOUN
app01-11088	82	6	with	with	ADP
app01-11088	82	7	a	a	DET
app01-11088	82	8	stride	stride	NOUN
app01-11088	82	9	of	of	ADP
app01-11088	82	10	2	2	NUM
app01-11088	82	11	and	and	CCONJ
app01-11088	82	12	kernels	kernel	NOUN
app01-11088	82	13	k	k	PROPN
app01-11088	82	14	of	of	ADP
app01-11088	82	15	size	size	NOUN
app01-11088	82	16	of	of	ADP
app01-11088	82	17	(	(	PUNCT
app01-11088	82	18	4,4	4,4	NUM
app01-11088	82	19	)	)	PUNCT
app01-11088	82	20	is	be	AUX
app01-11088	82	21	used	use	VERB
app01-11088	82	22	to	to	PART
app01-11088	82	23	up	up	ADP
app01-11088	82	24	-	-	PUNCT
app01-11088	82	25	sample	sample	VERB
app01-11088	82	26	the	the	DET
app01-11088	82	27	images	image	NOUN
app01-11088	82	28	to	to	ADP
app01-11088	82	29	twice	twice	DET
app01-11088	82	30	the	the	DET
app01-11088	82	31	spatial	spatial	ADJ
app01-11088	82	32	dimension	dimension	NOUN
app01-11088	82	33	of	of	ADP
app01-11088	82	34	the	the	DET
app01-11088	82	35	input	input	NOUN
app01-11088	82	36	,	,	PUNCT
app01-11088	82	37	while	while	SCONJ
app01-11088	82	38	the	the	DET
app01-11088	82	39	convolution	convolution	NOUN
app01-11088	82	40	layers	layer	NOUN
app01-11088	82	41	are	be	AUX
app01-11088	82	42	designed	design	VERB
app01-11088	82	43	to	to	PART
app01-11088	82	44	preserve	preserve	VERB
app01-11088	82	45	the	the	DET
app01-11088	82	46	size	size	NOUN
app01-11088	82	47	of	of	ADP
app01-11088	82	48	the	the	DET
app01-11088	82	49	output	output	NOUN
app01-11088	82	50	.	.	PUNCT
app01-11088	83	1	the	the	DET
app01-11088	83	2	depth	depth	NOUN
app01-11088	83	3	of	of	ADP
app01-11088	83	4	the	the	DET
app01-11088	83	5	decoder	decoder	NOUN
app01-11088	83	6	is	be	AUX
app01-11088	83	7	adjusted	adjust	VERB
app01-11088	83	8	such	such	ADJ
app01-11088	83	9	that	that	SCONJ
app01-11088	83	10	the	the	DET
app01-11088	83	11	input	input	NOUN
app01-11088	83	12	to	to	ADP
app01-11088	83	13	the	the	DET
app01-11088	83	14	decoder	decoder	NOUN
app01-11088	83	15	is	be	AUX
app01-11088	83	16	32	32	NUM
app01-11088	83	17	pixels	pixel	NOUN
app01-11088	83	18	in	in	ADP
app01-11088	83	19	both	both	DET
app01-11088	83	20	dimensions	dimension	NOUN
app01-11088	83	21	and	and	CCONJ
app01-11088	83	22	the	the	DET
app01-11088	83	23	output	output	NOUN
app01-11088	83	24	is	be	AUX
app01-11088	83	25	512	512	NUM
app01-11088	83	26	pixels	pixel	NOUN
app01-11088	83	27	in	in	ADP
app01-11088	83	28	both	both	DET
app01-11088	83	29	dimensions	dimension	NOUN
app01-11088	83	30	with	with	ADP
app01-11088	83	31	a	a	DET
app01-11088	83	32	feature	feature	NOUN
app01-11088	83	33	dimension	dimension	NOUN
app01-11088	83	34	of	of	ADP
app01-11088	83	35	size	size	NOUN
app01-11088	83	36	1	1	NUM
app01-11088	83	37	,	,	PUNCT
app01-11088	83	38	which	which	PRON
app01-11088	83	39	in	in	ADP
app01-11088	83	40	this	this	DET
app01-11088	83	41	case	case	NOUN
app01-11088	83	42	necessitates	necessitate	VERB
app01-11088	83	43	the	the	DET
app01-11088	83	44	use	use	NOUN
app01-11088	83	45	of	of	ADP
app01-11088	83	46	four	four	NUM
app01-11088	83	47	transposed	transpose	VERB
app01-11088	83	48	convolution	convolution	NOUN
app01-11088	83	49	layers	layer	NOUN
app01-11088	83	50	.	.	PUNCT
app01-11088	84	1	details	detail	NOUN
app01-11088	84	2	about	about	ADP
app01-11088	84	3	the	the	DET
app01-11088	84	4	decoder	decoder	NOUN
app01-11088	84	5	is	be	AUX
app01-11088	84	6	presented	present	VERB
app01-11088	84	7	in	in	ADP
app01-11088	84	8	table	table	NOUN
app01-11088	84	9	3	3	NUM
app01-11088	84	10	.	.	PUNCT
app01-11088	85	1	finally	finally	ADV
app01-11088	85	2	,	,	PUNCT
app01-11088	85	3	the	the	DET
app01-11088	85	4	resulting	result	VERB
app01-11088	85	5	model	model	NOUN
app01-11088	85	6	is	be	AUX
app01-11088	85	7	trained	train	VERB
app01-11088	85	8	to	to	PART
app01-11088	85	9	obtain	obtain	VERB
app01-11088	85	10	a	a	DET
app01-11088	85	11	set	set	NOUN
app01-11088	85	12	of	of	ADP
app01-11088	85	13	parameters	parameter	NOUN
app01-11088	85	14	θ	θ	PROPN
app01-11088	85	15	.	.	PUNCT
app01-11088	86	1	inferencing	inference	VERB
app01-11088	86	2	using	use	VERB
app01-11088	86	3	this	this	DET
app01-11088	86	4	model	model	NOUN
app01-11088	86	5	is	be	AUX
app01-11088	86	6	subsequently	subsequently	ADV
app01-11088	86	7	performed	perform	VERB
app01-11088	86	8	by	by	ADP
app01-11088	86	9	using	use	VERB
app01-11088	86	10	this	this	DET
app01-11088	86	11	fixed	fix	VERB
app01-11088	86	12	set	set	NOUN
app01-11088	86	13	of	of	ADP
app01-11088	86	14	parameters	parameter	NOUN
app01-11088	86	15	θ	θ	PROPN
app01-11088	86	16	such	such	ADJ
app01-11088	86	17	that	that	SCONJ
app01-11088	86	18	given	give	VERB
app01-11088	86	19	a	a	DET
app01-11088	86	20	pair	pair	NOUN
app01-11088	86	21	of	of	ADP
app01-11088	86	22	shape	shape	NOUN
app01-11088	86	23	constants	constant	NOUN
app01-11088	86	24	and	and	CCONJ
app01-11088	86	25	state	state	NOUN
app01-11088	86	26	descriptors	descriptor	NOUN
app01-11088	86	27	,	,	PUNCT
app01-11088	86	28	the	the	DET
app01-11088	86	29	model	model	NOUN
app01-11088	86	30	is	be	AUX
app01-11088	86	31	able	able	ADJ
app01-11088	86	32	to	to	PART
app01-11088	86	33	generate	generate	VERB
app01-11088	86	34	an	an	DET
app01-11088	86	35	approximate	approximate	ADJ
app01-11088	86	36	2d	2d	NUM
app01-11088	86	37	aggregate	aggregate	NOUN
app01-11088	86	38	indicated	indicate	VERB
app01-11088	86	39	by	by	ADP
app01-11088	86	40	these	these	DET
app01-11088	86	41	quantities	quantity	NOUN
app01-11088	86	42	.	.	PUNCT
app01-11088	87	1	the	the	DET
app01-11088	87	2	overall	overall	ADJ
app01-11088	87	3	workflow	workflow	NOUN
app01-11088	87	4	for	for	ADP
app01-11088	87	5	training	train	VERB
app01-11088	87	6	the	the	DET
app01-11088	87	7	model	model	NOUN
app01-11088	87	8	is	be	AUX
app01-11088	87	9	presented	present	VERB
app01-11088	87	10	in	in	ADP
app01-11088	87	11	figure	figure	NOUN
app01-11088	87	12	2	2	NUM
app01-11088	87	13	and	and	CCONJ
app01-11088	87	14	the	the	DET
app01-11088	87	15	training	training	NOUN
app01-11088	87	16	details	detail	NOUN
app01-11088	87	17	is	be	AUX
app01-11088	87	18	provided	provide	VERB
app01-11088	87	19	in	in	ADP
app01-11088	87	20	section	section	NOUN
app01-11088	87	21	2.4	2.4	NUM
app01-11088	87	22	.	.	PUNCT
app01-11088	88	1	2.3	2.3	NUM
app01-11088	88	2	.	.	PUNCT
app01-11088	89	1	aggregate	aggregate	VERB
app01-11088	89	2	dataset	dataset	VERB
app01-11088	89	3	the	the	DET
app01-11088	89	4	image	image	NOUN
app01-11088	89	5	dataset	dataset	NOUN
app01-11088	89	6	consisted	consist	VERB
app01-11088	89	7	of	of	ADP
app01-11088	89	8	binary	binary	ADJ
app01-11088	89	9	images	image	NOUN
app01-11088	89	10	,	,	PUNCT
app01-11088	89	11	each	each	PRON
app01-11088	89	12	containing	contain	VERB
app01-11088	89	13	a	a	DET
app01-11088	89	14	single	single	ADJ
app01-11088	89	15	aggregate	aggregate	NOUN
app01-11088	89	16	.	.	PUNCT
app01-11088	90	1	aggregates	aggregate	NOUN
app01-11088	90	2	were	be	AUX
app01-11088	90	3	represented	represent	VERB
app01-11088	90	4	with	with	ADP
app01-11088	90	5	a	a	DET
app01-11088	90	6	pixel	pixel	NOUN
app01-11088	90	7	value	value	NOUN
app01-11088	90	8	of	of	ADP
app01-11088	90	9	1	1	NUM
app01-11088	90	10	,	,	PUNCT
app01-11088	90	11	while	while	SCONJ
app01-11088	90	12	0	0	NUM
app01-11088	90	13	indicated	indicate	VERB
app01-11088	90	14	the	the	DET
app01-11088	90	15	background	background	NOUN
app01-11088	90	16	.	.	PUNCT
app01-11088	91	1	the	the	DET
app01-11088	91	2	image	image	NOUN
app01-11088	91	3	dimensions	dimension	NOUN
app01-11088	91	4	were	be	AUX
app01-11088	91	5	adjusted	adjust	VERB
app01-11088	91	6	to	to	ADP
app01-11088	91	7	the	the	DET
app01-11088	91	8	smallest	small	ADJ
app01-11088	91	9	power	power	NOUN
app01-11088	91	10	of	of	ADP
app01-11088	91	11	2	2	NUM
app01-11088	91	12	that	that	PRON
app01-11088	91	13	could	could	AUX
app01-11088	91	14	accommodate	accommodate	VERB
app01-11088	91	15	the	the	DET
app01-11088	91	16	largest	large	ADJ
app01-11088	91	17	aggregate	aggregate	NOUN
app01-11088	91	18	,	,	PUNCT
app01-11088	91	19	resulting	result	VERB
app01-11088	91	20	in	in	ADP
app01-11088	91	21	each	each	DET
app01-11088	91	22	image	image	NOUN
app01-11088	91	23	measuring	measure	VERB
app01-11088	91	24	512	512	NUM
app01-11088	91	25	×	×	NOUN
app01-11088	91	26	512	512	NUM
app01-11088	91	27	pixels	pixel	NOUN
app01-11088	91	28	.	.	PUNCT
app01-11088	92	1	each	each	DET
app01-11088	92	2	1	1	NUM
app01-11088	92	3	mm	mm	NOUN
app01-11088	92	4	of	of	ADP
app01-11088	92	5	spatial	spatial	ADJ
app01-11088	92	6	dimension	dimension	NOUN
app01-11088	92	7	corresponded	correspond	VERB
app01-11088	92	8	to	to	ADP
app01-11088	92	9	10.24	10.24	NUM
app01-11088	92	10	pixels	pixel	NOUN
app01-11088	92	11	.	.	PUNCT
app01-11088	93	1	examples	example	NOUN
app01-11088	93	2	of	of	ADP
app01-11088	93	3	training	training	NOUN
app01-11088	93	4	samples	sample	NOUN
app01-11088	93	5	are	be	AUX
app01-11088	93	6	shown	show	VERB
app01-11088	93	7	in	in	ADP
app01-11088	93	8	figure	figure	NOUN
app01-11088	93	9	3	3	NUM
app01-11088	93	10	.	.	PUNCT
app01-11088	93	11	training	training	NOUN
app01-11088	93	12	data	datum	NOUN
app01-11088	93	13	comprised	comprise	VERB
app01-11088	93	14	(	(	PUNCT
app01-11088	93	15	x	x	X
app01-11088	93	16	,	,	PUNCT
app01-11088	93	17	y	y	PROPN
app01-11088	93	18	)	)	PUNCT
app01-11088	93	19	pairs	pair	NOUN
app01-11088	93	20	,	,	PUNCT
app01-11088	93	21	where	where	SCONJ
app01-11088	93	22	x	x	PRON
app01-11088	93	23	represented	represent	VERB
app01-11088	93	24	the	the	DET
app01-11088	93	25	input	input	NOUN
app01-11088	93	26	data	datum	NOUN
app01-11088	93	27	and	and	CCONJ
app01-11088	93	28	y	y	PRON
app01-11088	93	29	the	the	DET
app01-11088	93	30	target	target	NOUN
app01-11088	93	31	.	.	PUNCT
app01-11088	94	1	each	each	PRON
app01-11088	94	2	x	x	NUM
app01-11088	94	3	consisted	consist	VERB
app01-11088	94	4	of	of	ADP
app01-11088	94	5	state	state	NOUN
app01-11088	94	6	descriptors	descriptor	NOUN
app01-11088	94	7	(	(	PUNCT
app01-11088	94	8	sd	sd	NOUN
app01-11088	94	9	)	)	PUNCT
app01-11088	94	10	and	and	CCONJ
app01-11088	94	11	shape	shape	NOUN
app01-11088	94	12	constants	constant	NOUN
app01-11088	94	13	(	(	PUNCT
app01-11088	94	14	sc	sc	PROPN
app01-11088	94	15	)	)	PUNCT
app01-11088	94	16	.	.	PUNCT
app01-11088	95	1	sd	sd	NOUN
app01-11088	95	2	corresponded	correspond	VERB
app01-11088	95	3	to	to	ADP
app01-11088	95	4	the	the	DET
app01-11088	95	5	area	area	NOUN
app01-11088	95	6	of	of	ADP
app01-11088	95	7	the	the	DET
app01-11088	95	8	aggregates	aggregate	NOUN
app01-11088	95	9	,	,	PUNCT
app01-11088	95	10	stop	stop	VERB
app01-11088	95	11	?	?	PUNCT
app01-11088	96	1	𝜽	𝜽	ADP
app01-11088	96	2	update	update	NOUN
app01-11088	96	3	parameters	parameter	NOUN
app01-11088	96	4	target	target	NOUN
app01-11088	96	5	shape	shape	NOUN
app01-11088	96	6	predicted	predict	VERB
app01-11088	96	7	shape	shape	NOUN
app01-11088	96	8	no	no	DET
app01-11088	96	9	yes	yes	INTJ
app01-11088	96	10	mse	mse	NOUN
app01-11088	96	11	loss	loss	NOUN
app01-11088	96	12	d	d	X
app01-11088	96	13	e	e	NOUN
app01-11088	96	14	s	s	PROPN
app01-11088	96	15	c	c	NOUN
app01-11088	96	16	r	r	NOUN
app01-11088	97	1	i	i	PRON
app01-11088	97	2	p	p	NOUN
app01-11088	98	1	t	t	X
app01-11088	98	2	o	o	X
app01-11088	98	3	r	r	NOUN
app01-11088	98	4	s	s	NOUN
app01-11088	98	5	figure	figure	NOUN
app01-11088	98	6	2	2	NUM
app01-11088	98	7	.	.	PUNCT
app01-11088	98	8	training	training	NOUN
app01-11088	98	9	workflow	workflow	NOUN
app01-11088	98	10	based	base	VERB
app01-11088	98	11	on	on	ADP
app01-11088	98	12	supervised	supervised	ADJ
app01-11088	98	13	learning	learning	NOUN
app01-11088	98	14	scheme	scheme	NOUN
app01-11088	98	15	.	.	PUNCT
app01-11088	99	1	figure	figure	NOUN
app01-11088	99	2	3	3	NUM
app01-11088	99	3	.	.	NOUN
app01-11088	99	4	example	example	NOUN
app01-11088	99	5	of	of	ADP
app01-11088	99	6	training	training	NOUN
app01-11088	99	7	dataset	dataset	NOUN
app01-11088	99	8	containing	contain	VERB
app01-11088	99	9	8	8	NUM
app01-11088	99	10	500	500	NUM
app01-11088	99	11	samples	sample	NOUN
app01-11088	99	12	.	.	PUNCT
app01-11088	100	1	represented	represent	VERB
app01-11088	100	2	by	by	ADP
app01-11088	100	3	m00	m00	NOUN
app01-11088	100	4	,	,	PUNCT
app01-11088	100	5	and	and	CCONJ
app01-11088	100	6	sc	sc	PROPN
app01-11088	100	7	comprised	comprise	VERB
app01-11088	100	8	normalized	normalize	VERB
app01-11088	100	9	central	central	ADJ
app01-11088	100	10	moments	moment	NOUN
app01-11088	100	11	,	,	PUNCT
app01-11088	100	12	η20	η20	NOUN
app01-11088	100	13	,	,	PUNCT
app01-11088	100	14	η11	η11	NOUN
app01-11088	100	15	,	,	PUNCT
app01-11088	100	16	·	·	PUNCT
app01-11088	100	17	·	·	PUNCT
app01-11088	101	1	·	·	PUNCT
app01-11088	101	2	,	,	PUNCT
app01-11088	101	3	up	up	ADP
app01-11088	101	4	to	to	ADP
app01-11088	101	5	the	the	DET
app01-11088	101	6	order	order	NOUN
app01-11088	101	7	(	(	PUNCT
app01-11088	101	8	p	p	X
app01-11088	101	9	+	+	NOUN
app01-11088	101	10	q	q	X
app01-11088	101	11	)	)	PUNCT
app01-11088	101	12	=	=	SYM
app01-11088	101	13	7	7	X
app01-11088	101	14	.	.	PUNCT
app01-11088	102	1	the	the	DET
app01-11088	102	2	target	target	NOUN
app01-11088	102	3	y	y	PROPN
app01-11088	102	4	was	be	AUX
app01-11088	102	5	the	the	DET
app01-11088	102	6	binary	binary	ADJ
app01-11088	102	7	image	image	NOUN
app01-11088	102	8	of	of	ADP
app01-11088	102	9	the	the	DET
app01-11088	102	10	aggregate	aggregate	NOUN
app01-11088	102	11	from	from	ADP
app01-11088	102	12	which	which	PRON
app01-11088	102	13	x	x	PRON
app01-11088	102	14	was	be	AUX
app01-11088	102	15	derived	derive	VERB
app01-11088	102	16	.	.	PUNCT
app01-11088	103	1	a	a	DET
app01-11088	103	2	total	total	NOUN
app01-11088	103	3	of	of	ADP
app01-11088	103	4	8	8	NUM
app01-11088	103	5	500	500	NUM
app01-11088	103	6	labeled	label	VERB
app01-11088	103	7	(	(	PUNCT
app01-11088	103	8	x	x	X
app01-11088	103	9	,	,	PUNCT
app01-11088	103	10	y	y	NOUN
app01-11088	103	11	)	)	PUNCT
app01-11088	103	12	pairs	pair	NOUN
app01-11088	103	13	were	be	AUX
app01-11088	103	14	used	use	VERB
app01-11088	103	15	for	for	ADP
app01-11088	103	16	training	training	NOUN
app01-11088	103	17	,	,	PUNCT
app01-11088	103	18	while	while	SCONJ
app01-11088	103	19	a	a	DET
app01-11088	103	20	separate	separate	ADJ
app01-11088	103	21	subset	subset	NOUN
app01-11088	103	22	of	of	ADP
app01-11088	103	23	samples	sample	NOUN
app01-11088	103	24	was	be	AUX
app01-11088	103	25	reserved	reserve	VERB
app01-11088	103	26	for	for	ADP
app01-11088	103	27	testing	testing	NOUN
app01-11088	103	28	and	and	CCONJ
app01-11088	103	29	excluded	exclude	VERB
app01-11088	103	30	from	from	ADP
app01-11088	103	31	training	training	NOUN
app01-11088	103	32	.	.	PUNCT
app01-11088	104	1	additionally	additionally	ADV
app01-11088	104	2	,	,	PUNCT
app01-11088	104	3	no	no	DET
app01-11088	104	4	data	data	NOUN
app01-11088	104	5	augmentation	augmentation	NOUN
app01-11088	104	6	was	be	AUX
app01-11088	104	7	applied	apply	VERB
app01-11088	104	8	to	to	PART
app01-11088	104	9	supplement	supplement	VERB
app01-11088	104	10	the	the	DET
app01-11088	104	11	original	original	ADJ
app01-11088	104	12	dataset	dataset	NOUN
app01-11088	104	13	.	.	PUNCT
app01-11088	105	1	2.4	2.4	NUM
app01-11088	105	2	.	.	PUNCT
app01-11088	106	1	training	train	VERB
app01-11088	106	2	the	the	DET
app01-11088	106	3	workflow	workflow	NOUN
app01-11088	106	4	was	be	AUX
app01-11088	106	5	implemented	implement	VERB
app01-11088	106	6	using	use	VERB
app01-11088	106	7	the	the	DET
app01-11088	106	8	pytorch	pytorch	NOUN
app01-11088	106	9	framework	framework	NOUN
app01-11088	106	10	[	[	X
app01-11088	106	11	18	18	NUM
app01-11088	106	12	]	]	PUNCT
app01-11088	106	13	.	.	PUNCT
app01-11088	107	1	the	the	DET
app01-11088	107	2	model	model	NOUN
app01-11088	107	3	was	be	AUX
app01-11088	107	4	trained	train	VERB
app01-11088	107	5	with	with	ADP
app01-11088	107	6	the	the	DET
app01-11088	107	7	adam	adam	PROPN
app01-11088	107	8	optimizer	optimizer	NOUN
app01-11088	107	9	[	[	X
app01-11088	107	10	19	19	NUM
app01-11088	107	11	]	]	PUNCT
app01-11088	107	12	at	at	ADP
app01-11088	107	13	a	a	DET
app01-11088	107	14	learning	learning	NOUN
app01-11088	107	15	rate	rate	NOUN
app01-11088	107	16	of	of	ADP
app01-11088	107	17	0.001	0.001	NUM
app01-11088	107	18	,	,	PUNCT
app01-11088	107	19	with	with	ADP
app01-11088	107	20	cosine	cosine	NOUN
app01-11088	107	21	annealing	annealing	NOUN
app01-11088	107	22	applied	apply	VERB
app01-11088	107	23	for	for	ADP
app01-11088	107	24	learning	learn	VERB
app01-11088	107	25	rate	rate	NOUN
app01-11088	107	26	scheduling	scheduling	NOUN
app01-11088	107	27	over	over	ADP
app01-11088	107	28	96	96	NUM
app01-11088	107	29	epochs	epoch	NOUN
app01-11088	107	30	.	.	PUNCT
app01-11088	108	1	the	the	DET
app01-11088	108	2	mean	mean	ADJ
app01-11088	108	3	squared	square	VERB
app01-11088	108	4	error	error	NOUN
app01-11088	108	5	was	be	AUX
app01-11088	108	6	calculated	calculate	VERB
app01-11088	108	7	between	between	ADP
app01-11088	108	8	the	the	DET
app01-11088	108	9	model	model	NOUN
app01-11088	108	10	’s	’s	PART
app01-11088	108	11	output	output	NOUN
app01-11088	108	12	and	and	CCONJ
app01-11088	108	13	the	the	DET
app01-11088	108	14	target	target	NOUN
app01-11088	108	15	image	image	NOUN
app01-11088	108	16	as	as	ADP
app01-11088	108	17	the	the	DET
app01-11088	108	18	loss	loss	NOUN
app01-11088	108	19	function	function	NOUN
app01-11088	108	20	.	.	PUNCT
app01-11088	109	1	training	training	NOUN
app01-11088	109	2	was	be	AUX
app01-11088	109	3	conducted	conduct	VERB
app01-11088	109	4	using	use	VERB
app01-11088	109	5	minibatches	minibatche	NOUN
app01-11088	109	6	of	of	ADP
app01-11088	109	7	20	20	NUM
app01-11088	109	8	samples	sample	NOUN
app01-11088	109	9	.	.	PUNCT
app01-11088	110	1	hyperparameters	hyperparameter	NOUN
app01-11088	110	2	,	,	PUNCT
app01-11088	110	3	including	include	VERB
app01-11088	110	4	the	the	DET
app01-11088	110	5	learning	learning	NOUN
app01-11088	110	6	rate	rate	NOUN
app01-11088	110	7	,	,	PUNCT
app01-11088	110	8	batch	batch	NOUN
app01-11088	110	9	size	size	NOUN
app01-11088	110	10	,	,	PUNCT
app01-11088	110	11	and	and	CCONJ
app01-11088	110	12	optimizer	optimizer	NOUN
app01-11088	110	13	settings	setting	NOUN
app01-11088	110	14	,	,	PUNCT
app01-11088	110	15	were	be	AUX
app01-11088	110	16	experimentally	experimentally	ADV
app01-11088	110	17	determined	determine	VERB
app01-11088	110	18	to	to	PART
app01-11088	110	19	achieve	achieve	VERB
app01-11088	110	20	optimal	optimal	ADJ
app01-11088	110	21	performance	performance	NOUN
app01-11088	110	22	.	.	PUNCT
app01-11088	111	1	the	the	DET
app01-11088	111	2	entire	entire	ADJ
app01-11088	111	3	training	training	NOUN
app01-11088	111	4	process	process	NOUN
app01-11088	111	5	took	take	VERB
app01-11088	111	6	approximately	approximately	ADV
app01-11088	111	7	2	2	NUM
app01-11088	111	8	hours	hour	NOUN
app01-11088	111	9	on	on	ADP
app01-11088	111	10	a	a	DET
app01-11088	111	11	workstation	workstation	NOUN
app01-11088	111	12	equipped	equip	VERB
app01-11088	111	13	with	with	ADP
app01-11088	111	14	an	an	DET
app01-11088	111	15	nvidia	nvidia	PROPN
app01-11088	111	16	rtx	rtx	PROPN
app01-11088	111	17	4090	4090	NUM
app01-11088	111	18	gpu	gpu	PROPN
app01-11088	111	19	and	and	CCONJ
app01-11088	111	20	an	an	DET
app01-11088	111	21	amd	amd	ADJ
app01-11088	111	22	ryzen9	ryzen9	PROPN
app01-11088	111	23	7950x	7950x	NOUN
app01-11088	111	24	cpu	cpu	NOUN
app01-11088	111	25	.	.	PUNCT
app01-11088	112	1	inferencing	inference	VERB
app01-11088	112	2	with	with	ADP
app01-11088	112	3	the	the	DET
app01-11088	112	4	trained	train	VERB
app01-11088	112	5	model	model	NOUN
app01-11088	112	6	typically	typically	ADV
app01-11088	112	7	required	require	VERB
app01-11088	112	8	about	about	ADV
app01-11088	112	9	1	1	NUM
app01-11088	112	10	minute	minute	NOUN
app01-11088	112	11	for	for	ADP
app01-11088	112	12	approximately	approximately	ADV
app01-11088	112	13	440	440	NUM
app01-11088	112	14	samples	sample	NOUN
app01-11088	112	15	.	.	PUNCT
app01-11088	113	1	3	3	X
app01-11088	113	2	.	.	X
app01-11088	113	3	results	result	NOUN
app01-11088	113	4	for	for	ADP
app01-11088	113	5	the	the	DET
app01-11088	113	6	results	result	NOUN
app01-11088	113	7	presented	present	VERB
app01-11088	113	8	in	in	ADP
app01-11088	113	9	figure	figure	NOUN
app01-11088	113	10	4	4	NUM
app01-11088	113	11	,	,	PUNCT
app01-11088	113	12	the	the	DET
app01-11088	113	13	neural	neural	ADJ
app01-11088	113	14	network	network	NOUN
app01-11088	113	15	was	be	AUX
app01-11088	113	16	applied	apply	VERB
app01-11088	113	17	to	to	PART
app01-11088	113	18	reconstruct	reconstruct	VERB
app01-11088	113	19	the	the	DET
app01-11088	113	20	aggregates	aggregate	NOUN
app01-11088	113	21	us19	us19	PROPN
app01-11088	113	22	k.	k.	PUNCT
app01-11088	113	23	das	das	PROPN
app01-11088	113	24	,	,	PUNCT
app01-11088	113	25	j.	j.	PROPN
app01-11088	113	26	sýkora	sýkora	PROPN
app01-11088	113	27	,	,	PUNCT
app01-11088	113	28	a.	a.	PROPN
app01-11088	113	29	kučerová	kučerová	PROPN
app01-11088	113	30	acta	acta	PROPN
app01-11088	113	31	polytechnica	polytechnica	PROPN
app01-11088	113	32	ctu	ctu	NOUN
app01-11088	113	33	proceedings	proceeding	NOUN
app01-11088	113	34	layer	layer	NOUN
app01-11088	113	35	in	in	ADP
app01-11088	113	36	/	/	SYM
app01-11088	113	37	out	out	ADP
app01-11088	113	38	-	-	PUNCT
app01-11088	113	39	channels	channel	NOUN
app01-11088	113	40	kernel	kernel	NOUN
app01-11088	113	41	stride	stride	NOUN
app01-11088	113	42	padding	padding	NOUN
app01-11088	113	43	activation	activation	NOUN
app01-11088	113	44	function	function	VERB
app01-11088	113	45	transposedconv2d	transposedconv2d	ADJ
app01-11088	113	46	2/512	2/512	NUM
app01-11088	113	47	(	(	PUNCT
app01-11088	113	48	4	4	NUM
app01-11088	113	49	×	×	NOUN
app01-11088	113	50	4	4	NUM
app01-11088	113	51	)	)	PUNCT
app01-11088	113	52	2	2	NUM
app01-11088	113	53	1	1	NUM
app01-11088	114	1	relu	relu	NOUN
app01-11088	114	2	conv2d	conv2d	NOUN
app01-11088	114	3	512/512	512/512	NUM
app01-11088	114	4	(	(	PUNCT
app01-11088	114	5	3	3	NUM
app01-11088	114	6	×	×	NOUN
app01-11088	114	7	3	3	NUM
app01-11088	114	8	)	)	PUNCT
app01-11088	114	9	1	1	NUM
app01-11088	114	10	1	1	NUM
app01-11088	114	11	relu	relu	NOUN
app01-11088	114	12	transposedconv2d	transposedconv2d	PROPN
app01-11088	115	1	512/256	512/256	NUM
app01-11088	115	2	(	(	PUNCT
app01-11088	115	3	4	4	NUM
app01-11088	115	4	×	×	NOUN
app01-11088	115	5	4	4	NUM
app01-11088	115	6	)	)	PUNCT
app01-11088	115	7	2	2	NUM
app01-11088	115	8	1	1	NUM
app01-11088	115	9	relu	relu	NOUN
app01-11088	115	10	conv2d	conv2d	VERB
app01-11088	115	11	256/256	256/256	NUM
app01-11088	115	12	(	(	PUNCT
app01-11088	115	13	3	3	NUM
app01-11088	115	14	×	×	NOUN
app01-11088	115	15	3	3	NUM
app01-11088	115	16	)	)	PUNCT
app01-11088	115	17	1	1	NUM
app01-11088	115	18	1	1	NUM
app01-11088	116	1	relu	relu	NOUN
app01-11088	116	2	transposedconv2d	transposedconv2d	PROPN
app01-11088	117	1	256/128	256/128	PROPN
app01-11088	117	2	(	(	PUNCT
app01-11088	117	3	4	4	NUM
app01-11088	117	4	×	×	NOUN
app01-11088	117	5	4	4	NUM
app01-11088	117	6	)	)	PUNCT
app01-11088	117	7	2	2	NUM
app01-11088	117	8	1	1	NUM
app01-11088	117	9	relu	relu	NOUN
app01-11088	117	10	conv2d	conv2d	VERB
app01-11088	117	11	128/128	128/128	NUM
app01-11088	117	12	(	(	PUNCT
app01-11088	117	13	3	3	NUM
app01-11088	117	14	×	×	NOUN
app01-11088	117	15	3	3	NUM
app01-11088	117	16	)	)	PUNCT
app01-11088	117	17	1	1	NUM
app01-11088	117	18	1	1	NUM
app01-11088	117	19	relu	relu	NOUN
app01-11088	117	20	transposedconv2d	transposedconv2d	PROPN
app01-11088	117	21	128/64	128/64	NUM
app01-11088	117	22	(	(	PUNCT
app01-11088	117	23	4	4	NUM
app01-11088	117	24	×	×	NOUN
app01-11088	117	25	4	4	NUM
app01-11088	117	26	)	)	PUNCT
app01-11088	117	27	2	2	NUM
app01-11088	117	28	1	1	NUM
app01-11088	117	29	relu	relu	NOUN
app01-11088	117	30	conv2d	conv2d	VERB
app01-11088	118	1	64/64	64/64	NUM
app01-11088	118	2	(	(	PUNCT
app01-11088	118	3	3	3	NUM
app01-11088	118	4	×	×	NOUN
app01-11088	118	5	3	3	NUM
app01-11088	118	6	)	)	PUNCT
app01-11088	118	7	1	1	NUM
app01-11088	118	8	1	1	NUM
app01-11088	118	9	relu	relu	NOUN
app01-11088	118	10	conv2d	conv2d	VERB
app01-11088	118	11	64/64	64/64	NUM
app01-11088	118	12	(	(	PUNCT
app01-11088	118	13	1	1	NUM
app01-11088	118	14	×	×	NOUN
app01-11088	118	15	1	1	NUM
app01-11088	118	16	)	)	PUNCT
app01-11088	118	17	1	1	NUM
app01-11088	118	18	0	0	NUM
app01-11088	118	19	relu	relu	NOUN
app01-11088	118	20	table	table	NOUN
app01-11088	118	21	3	3	NUM
app01-11088	118	22	.	.	PUNCT
app01-11088	118	23	neural	neural	ADJ
app01-11088	118	24	network	network	NOUN
app01-11088	118	25	architecture	architecture	NOUN
app01-11088	118	26	of	of	ADP
app01-11088	118	27	decoder	decoder	NOUN
app01-11088	118	28	with	with	ADP
app01-11088	118	29	corresponding	correspond	VERB
app01-11088	118	30	convolutional	convolutional	ADJ
app01-11088	118	31	layers	layer	NOUN
app01-11088	118	32	.	.	PUNCT
app01-11088	119	1	original	original	ADJ
app01-11088	119	2	generated	generate	VERB
app01-11088	119	3	figure	figure	NOUN
app01-11088	119	4	4	4	NUM
app01-11088	119	5	.	.	PUNCT
app01-11088	119	6	reconstruction	reconstruction	NOUN
app01-11088	119	7	of	of	ADP
app01-11088	119	8	original	original	ADJ
app01-11088	119	9	aggregates	aggregate	NOUN
app01-11088	119	10	.	.	PUNCT
app01-11088	120	1	original	original	ADJ
app01-11088	120	2	generated	generate	VERB
app01-11088	120	3	0.25x	0.25x	PROPN
app01-11088	120	4	generated	generate	VERB
app01-11088	120	5	0.75x	0.75x	PROPN
app01-11088	120	6	generated	generate	VERB
app01-11088	120	7	1.25x	1.25x	NUM
app01-11088	120	8	generated	generate	VERB
app01-11088	120	9	1.75x	1.75x	NUM
app01-11088	120	10	figure	figure	NOUN
app01-11088	120	11	5	5	NUM
app01-11088	120	12	.	.	PUNCT
app01-11088	120	13	microstructure	microstructure	ADJ
app01-11088	120	14	generator	generator	NOUN
app01-11088	120	15	of	of	ADP
app01-11088	120	16	aggregates	aggregate	NOUN
app01-11088	120	17	based	base	VERB
app01-11088	120	18	on	on	ADP
app01-11088	120	19	the	the	DET
app01-11088	120	20	scaled	scale	VERB
app01-11088	120	21	area	area	NOUN
app01-11088	120	22	of	of	ADP
app01-11088	120	23	the	the	DET
app01-11088	120	24	original	original	ADJ
app01-11088	120	25	structure	structure	NOUN
app01-11088	120	26	.	.	PUNCT
app01-11088	121	1	ing	ing	ADJ
app01-11088	121	2	geometric	geometric	ADJ
app01-11088	121	3	descriptors	descriptor	NOUN
app01-11088	121	4	obtained	obtain	VERB
app01-11088	121	5	from	from	ADP
app01-11088	121	6	the	the	DET
app01-11088	121	7	original	original	ADJ
app01-11088	121	8	images	image	NOUN
app01-11088	121	9	.	.	PUNCT
app01-11088	122	1	it	it	PRON
app01-11088	122	2	is	be	AUX
app01-11088	122	3	evident	evident	ADJ
app01-11088	122	4	that	that	SCONJ
app01-11088	122	5	the	the	DET
app01-11088	122	6	model	model	NOUN
app01-11088	122	7	can	can	AUX
app01-11088	122	8	successfully	successfully	ADV
app01-11088	122	9	reconstruct	reconstruct	VERB
app01-11088	122	10	the	the	DET
app01-11088	122	11	aggregate	aggregate	ADJ
app01-11088	122	12	shapes	shape	NOUN
app01-11088	122	13	using	use	VERB
app01-11088	122	14	only	only	ADV
app01-11088	122	15	the	the	DET
app01-11088	122	16	34	34	NUM
app01-11088	122	17	quantities	quantity	NOUN
app01-11088	122	18	.	.	PUNCT
app01-11088	123	1	next	next	ADJ
app01-11088	123	2	,	,	PUNCT
app01-11088	123	3	to	to	PART
app01-11088	123	4	test	test	VERB
app01-11088	123	5	whether	whether	SCONJ
app01-11088	123	6	the	the	DET
app01-11088	123	7	model	model	NOUN
app01-11088	123	8	could	could	AUX
app01-11088	123	9	recognize	recognize	VERB
app01-11088	123	10	the	the	DET
app01-11088	123	11	scale	scale	NOUN
app01-11088	123	12	-	-	PUNCT
app01-11088	123	13	invariant	invariant	ADJ
app01-11088	123	14	property	property	NOUN
app01-11088	123	15	of	of	ADP
app01-11088	123	16	the	the	DET
app01-11088	123	17	normalized	normalize	VERB
app01-11088	123	18	central	central	ADJ
app01-11088	123	19	moments	moment	NOUN
app01-11088	123	20	,	,	PUNCT
app01-11088	123	21	the	the	DET
app01-11088	123	22	model	model	NOUN
app01-11088	123	23	was	be	AUX
app01-11088	123	24	applied	apply	VERB
app01-11088	123	25	to	to	ADP
app01-11088	123	26	the	the	DET
app01-11088	123	27	same	same	ADJ
app01-11088	123	28	set	set	NOUN
app01-11088	123	29	of	of	ADP
app01-11088	123	30	descriptors	descriptor	NOUN
app01-11088	123	31	where	where	SCONJ
app01-11088	123	32	only	only	ADV
app01-11088	123	33	the	the	DET
app01-11088	123	34	area	area	NOUN
app01-11088	123	35	,	,	PUNCT
app01-11088	123	36	i.e.	i.e.	X
app01-11088	123	37	m00	m00	PROPN
app01-11088	123	38	was	be	AUX
app01-11088	123	39	scaled	scale	VERB
app01-11088	123	40	by	by	ADP
app01-11088	123	41	a	a	DET
app01-11088	123	42	factor	factor	NOUN
app01-11088	123	43	ranging	range	VERB
app01-11088	123	44	from	from	ADP
app01-11088	123	45	0.25	0.25	NUM
app01-11088	123	46	to	to	ADP
app01-11088	123	47	2.0	2.0	NUM
app01-11088	123	48	in	in	ADP
app01-11088	123	49	increments	increment	NOUN
app01-11088	123	50	of	of	ADP
app01-11088	123	51	0.25	0.25	NUM
app01-11088	123	52	.	.	PUNCT
app01-11088	124	1	the	the	DET
app01-11088	124	2	results	result	NOUN
app01-11088	124	3	are	be	AUX
app01-11088	124	4	presented	present	VERB
app01-11088	124	5	in	in	ADP
app01-11088	124	6	figure	figure	NOUN
app01-11088	124	7	5	5	NUM
app01-11088	124	8	.	.	PUNCT
app01-11088	125	1	it	it	PRON
app01-11088	125	2	is	be	AUX
app01-11088	125	3	clear	clear	ADJ
app01-11088	125	4	that	that	SCONJ
app01-11088	125	5	the	the	DET
app01-11088	125	6	model	model	NOUN
app01-11088	125	7	can	can	AUX
app01-11088	125	8	recognize	recognize	VERB
app01-11088	125	9	the	the	DET
app01-11088	125	10	invariance	invariance	NOUN
app01-11088	125	11	between	between	ADP
app01-11088	125	12	aggregate	aggregate	ADJ
app01-11088	125	13	size	size	NOUN
app01-11088	125	14	and	and	CCONJ
app01-11088	125	15	its	its	PRON
app01-11088	125	16	moments	moment	NOUN
app01-11088	125	17	.	.	PUNCT
app01-11088	126	1	notably	notably	ADV
app01-11088	126	2	,	,	PUNCT
app01-11088	126	3	the	the	DET
app01-11088	126	4	dataset	dataset	NOUN
app01-11088	126	5	was	be	AUX
app01-11088	126	6	not	not	PART
app01-11088	126	7	augmented	augment	VERB
app01-11088	126	8	to	to	PART
app01-11088	126	9	account	account	VERB
app01-11088	126	10	for	for	ADP
app01-11088	126	11	the	the	DET
app01-11088	126	12	scaling	scale	VERB
app01-11088	126	13	invariance	invariance	NOUN
app01-11088	126	14	of	of	ADP
app01-11088	126	15	the	the	DET
app01-11088	126	16	moments	moment	NOUN
app01-11088	126	17	,	,	PUNCT
app01-11088	126	18	meaning	mean	VERB
app01-11088	126	19	this	this	DET
app01-11088	126	20	property	property	NOUN
app01-11088	126	21	was	be	AUX
app01-11088	126	22	implicitly	implicitly	ADV
app01-11088	126	23	learned	learn	VERB
app01-11088	126	24	by	by	ADP
app01-11088	126	25	the	the	DET
app01-11088	126	26	model	model	NOUN
app01-11088	126	27	.	.	PUNCT
app01-11088	127	1	the	the	DET
app01-11088	127	2	final	final	ADJ
app01-11088	127	3	results	result	NOUN
app01-11088	127	4	presented	present	VERB
app01-11088	127	5	in	in	ADP
app01-11088	127	6	figures	figure	NOUN
app01-11088	127	7	4	4	NUM
app01-11088	127	8	and	and	CCONJ
app01-11088	127	9	5	5	NUM
app01-11088	127	10	were	be	AUX
app01-11088	127	11	generated	generate	VERB
app01-11088	127	12	by	by	ADP
app01-11088	127	13	applying	apply	VERB
app01-11088	127	14	the	the	DET
app01-11088	127	15	neural	neural	ADJ
app01-11088	127	16	network	network	NOUN
app01-11088	127	17	to	to	ADP
app01-11088	127	18	a	a	DET
app01-11088	127	19	set	set	NOUN
app01-11088	127	20	of	of	ADP
app01-11088	127	21	descriptors	descriptor	NOUN
app01-11088	127	22	that	that	PRON
app01-11088	127	23	were	be	AUX
app01-11088	127	24	not	not	PART
app01-11088	127	25	previously	previously	ADV
app01-11088	127	26	used	use	VERB
app01-11088	127	27	for	for	ADP
app01-11088	127	28	training	training	NOUN
app01-11088	127	29	.	.	PUNCT
app01-11088	128	1	due	due	ADP
app01-11088	128	2	to	to	ADP
app01-11088	128	3	the	the	DET
app01-11088	128	4	requirement	requirement	NOUN
app01-11088	128	5	of	of	ADP
app01-11088	128	6	backpropagation	backpropagation	NOUN
app01-11088	128	7	during	during	ADP
app01-11088	128	8	training	training	NOUN
app01-11088	128	9	,	,	PUNCT
app01-11088	128	10	the	the	DET
app01-11088	128	11	output	output	NOUN
app01-11088	128	12	of	of	ADP
app01-11088	128	13	the	the	DET
app01-11088	128	14	neural	neural	ADJ
app01-11088	128	15	network	network	NOUN
app01-11088	128	16	is	be	AUX
app01-11088	128	17	a	a	DET
app01-11088	128	18	grayscale	grayscale	NOUN
app01-11088	128	19	image	image	NOUN
app01-11088	128	20	,	,	PUNCT
app01-11088	128	21	with	with	ADP
app01-11088	128	22	high	high	ADJ
app01-11088	128	23	pixel	pixel	NOUN
app01-11088	128	24	values	value	NOUN
app01-11088	128	25	indicating	indicate	VERB
app01-11088	128	26	aggregate	aggregate	ADJ
app01-11088	128	27	and	and	CCONJ
app01-11088	128	28	low	low	ADJ
app01-11088	128	29	pixel	pixel	NOUN
app01-11088	128	30	values	value	NOUN
app01-11088	128	31	indicating	indicate	VERB
app01-11088	128	32	their	their	PRON
app01-11088	128	33	absence	absence	NOUN
app01-11088	128	34	.	.	PUNCT
app01-11088	129	1	to	to	PART
app01-11088	129	2	obtain	obtain	VERB
app01-11088	129	3	the	the	DET
app01-11088	129	4	final	final	ADJ
app01-11088	129	5	binarized	binarize	VERB
app01-11088	129	6	result	result	NOUN
app01-11088	129	7	,	,	PUNCT
app01-11088	129	8	the	the	DET
app01-11088	129	9	model	model	NOUN
app01-11088	129	10	outputs	output	NOUN
app01-11088	129	11	were	be	AUX
app01-11088	129	12	thresholded	thresholde	VERB
app01-11088	129	13	using	use	VERB
app01-11088	129	14	otsu	otsu	PROPN
app01-11088	129	15	’s	’s	PART
app01-11088	129	16	method	method	NOUN
app01-11088	129	17	,	,	PUNCT
app01-11088	129	18	which	which	PRON
app01-11088	129	19	is	be	AUX
app01-11088	129	20	readily	readily	ADV
app01-11088	129	21	implemented	implement	VERB
app01-11088	129	22	in	in	ADP
app01-11088	129	23	the	the	DET
app01-11088	129	24	scikit	scikit	NOUN
app01-11088	129	25	-	-	PUNCT
app01-11088	129	26	image	image	NOUN
app01-11088	129	27	library	library	NOUN
app01-11088	129	28	[	[	X
app01-11088	129	29	20	20	NUM
app01-11088	129	30	]	]	PUNCT
app01-11088	129	31	.	.	PUNCT
app01-11088	130	1	the	the	DET
app01-11088	130	2	quality	quality	NOUN
app01-11088	130	3	of	of	ADP
app01-11088	130	4	the	the	DET
app01-11088	130	5	generated	generate	VERB
app01-11088	130	6	aggregates	aggregate	NOUN
app01-11088	130	7	was	be	AUX
app01-11088	130	8	assessed	assess	VERB
app01-11088	130	9	by	by	ADP
app01-11088	130	10	comparing	compare	VERB
app01-11088	130	11	the	the	DET
app01-11088	130	12	moments	moment	NOUN
app01-11088	130	13	of	of	ADP
app01-11088	130	14	the	the	DET
app01-11088	130	15	original	original	ADJ
app01-11088	130	16	aggregates	aggregate	NOUN
app01-11088	130	17	to	to	ADP
app01-11088	130	18	those	those	PRON
app01-11088	130	19	of	of	ADP
app01-11088	130	20	the	the	DET
app01-11088	130	21	generated	generate	VERB
app01-11088	130	22	aggregates	aggregate	NOUN
app01-11088	130	23	.	.	PUNCT
app01-11088	131	1	these	these	DET
app01-11088	131	2	results	result	NOUN
app01-11088	131	3	are	be	AUX
app01-11088	131	4	shown	show	VERB
app01-11088	131	5	in	in	ADP
app01-11088	131	6	figure	figure	NOUN
app01-11088	131	7	6	6	NUM
app01-11088	131	8	for	for	ADP
app01-11088	131	9	the	the	DET
app01-11088	131	10	various	various	ADJ
app01-11088	131	11	scaling	scale	VERB
app01-11088	131	12	factors	factor	NOUN
app01-11088	131	13	that	that	PRON
app01-11088	131	14	were	be	AUX
app01-11088	131	15	used	use	VERB
app01-11088	131	16	.	.	PUNCT
app01-11088	132	1	the	the	DET
app01-11088	132	2	scaling	scaling	NOUN
app01-11088	132	3	error	error	NOUN
app01-11088	132	4	is	be	AUX
app01-11088	132	5	presented	present	VERB
app01-11088	132	6	in	in	ADP
app01-11088	132	7	figure	figure	NOUN
app01-11088	132	8	7	7	NUM
app01-11088	132	9	,	,	PUNCT
app01-11088	132	10	20	20	NUM
app01-11088	132	11	vol	vol	NOUN
app01-11088	132	12	.	.	PUNCT
app01-11088	133	1	54/2025	54/2025	NUM
app01-11088	133	2	tractable	tractable	ADJ
app01-11088	133	3	descriptors	descriptor	NOUN
app01-11088	133	4	for	for	ADP
app01-11088	133	5	digital	digital	ADJ
app01-11088	133	6	aggregate	aggregate	ADJ
app01-11088	133	7	generation	generation	NOUN
app01-11088	133	8	figure	figure	NOUN
app01-11088	133	9	6	6	NUM
app01-11088	133	10	.	.	PUNCT
app01-11088	133	11	model	model	NOUN
app01-11088	133	12	performance	performance	NOUN
app01-11088	133	13	for	for	ADP
app01-11088	133	14	the	the	DET
app01-11088	133	15	various	various	ADJ
app01-11088	133	16	scaling	scale	VERB
app01-11088	133	17	factors	factor	NOUN
app01-11088	133	18	evaluated	evaluate	VERB
app01-11088	133	19	as	as	ADP
app01-11088	133	20	a	a	DET
app01-11088	133	21	mean	mean	ADJ
app01-11088	133	22	squared	square	VERB
app01-11088	133	23	error	error	NOUN
app01-11088	133	24	(	(	PUNCT
app01-11088	133	25	mse	mse	NOUN
app01-11088	133	26	)	)	PUNCT
app01-11088	133	27	of	of	ADP
app01-11088	133	28	normalized	normalize	VERB
app01-11088	133	29	central	central	ADJ
app01-11088	133	30	moments	moment	NOUN
app01-11088	133	31	.	.	PUNCT
app01-11088	134	1	figure	figure	VERB
app01-11088	134	2	7	7	NUM
app01-11088	134	3	.	.	PUNCT
app01-11088	134	4	scaling	scale	VERB
app01-11088	134	5	error	error	NOUN
app01-11088	134	6	analysis	analysis	NOUN
app01-11088	134	7	showing	show	VERB
app01-11088	134	8	the	the	DET
app01-11088	134	9	model	model	NOUN
app01-11088	134	10	’s	’s	PART
app01-11088	134	11	tendency	tendency	NOUN
app01-11088	134	12	to	to	PART
app01-11088	134	13	generate	generate	VERB
app01-11088	134	14	larger	large	ADJ
app01-11088	134	15	aggregates	aggregate	NOUN
app01-11088	134	16	for	for	ADP
app01-11088	134	17	scaling	scale	VERB
app01-11088	134	18	factors	factor	NOUN
app01-11088	134	19	less	less	ADJ
app01-11088	134	20	than	than	ADP
app01-11088	134	21	1	1	NUM
app01-11088	134	22	and	and	CCONJ
app01-11088	134	23	vice	vice	NOUN
app01-11088	134	24	versa	versa	ADV
app01-11088	134	25	.	.	PUNCT
app01-11088	135	1	where	where	SCONJ
app01-11088	135	2	the	the	DET
app01-11088	135	3	area	area	NOUN
app01-11088	135	4	of	of	ADP
app01-11088	135	5	the	the	DET
app01-11088	135	6	generated	generate	VERB
app01-11088	135	7	aggregates	aggregate	NOUN
app01-11088	135	8	is	be	AUX
app01-11088	135	9	compared	compare	VERB
app01-11088	135	10	to	to	ADP
app01-11088	135	11	the	the	DET
app01-11088	135	12	correct	correct	ADJ
app01-11088	135	13	area	area	NOUN
app01-11088	135	14	upon	upon	SCONJ
app01-11088	135	15	scaling	scale	VERB
app01-11088	135	16	.	.	PUNCT
app01-11088	136	1	it	it	PRON
app01-11088	136	2	is	be	AUX
app01-11088	136	3	evident	evident	ADJ
app01-11088	136	4	that	that	SCONJ
app01-11088	136	5	the	the	DET
app01-11088	136	6	model	model	NOUN
app01-11088	136	7	tends	tend	VERB
app01-11088	136	8	to	to	PART
app01-11088	136	9	generate	generate	VERB
app01-11088	136	10	larger	large	ADJ
app01-11088	136	11	aggregates	aggregate	NOUN
app01-11088	136	12	for	for	ADP
app01-11088	136	13	scaling	scale	VERB
app01-11088	136	14	factors	factor	NOUN
app01-11088	136	15	<	<	X
app01-11088	136	16	1	1	NUM
app01-11088	136	17	,	,	PUNCT
app01-11088	136	18	while	while	SCONJ
app01-11088	136	19	for	for	ADP
app01-11088	136	20	scaling	scale	VERB
app01-11088	136	21	factors	factor	NOUN
app01-11088	136	22	>	>	X
app01-11088	136	23	1	1	NUM
app01-11088	136	24	,	,	PUNCT
app01-11088	136	25	it	it	PRON
app01-11088	136	26	produces	produce	VERB
app01-11088	136	27	aggregates	aggregate	NOUN
app01-11088	136	28	with	with	ADP
app01-11088	136	29	smaller	small	ADJ
app01-11088	136	30	-	-	PUNCT
app01-11088	136	31	than	than	ADP
app01-11088	136	32	-	-	PUNCT
app01-11088	136	33	correct	correct	ADJ
app01-11088	136	34	areas	area	NOUN
app01-11088	136	35	.	.	PUNCT
app01-11088	137	1	this	this	PRON
app01-11088	137	2	may	may	AUX
app01-11088	137	3	be	be	AUX
app01-11088	137	4	attributed	attribute	VERB
app01-11088	137	5	to	to	ADP
app01-11088	137	6	the	the	DET
app01-11088	137	7	lack	lack	NOUN
app01-11088	137	8	of	of	ADP
app01-11088	137	9	explicit	explicit	ADJ
app01-11088	137	10	training	training	NOUN
app01-11088	137	11	on	on	ADP
app01-11088	137	12	a	a	DET
app01-11088	137	13	dataset	dataset	NOUN
app01-11088	137	14	augmented	augment	VERB
app01-11088	137	15	for	for	ADP
app01-11088	137	16	scale	scale	NOUN
app01-11088	137	17	invariance	invariance	NOUN
app01-11088	137	18	of	of	ADP
app01-11088	137	19	the	the	DET
app01-11088	137	20	moments	moment	NOUN
app01-11088	137	21	.	.	PUNCT
app01-11088	138	1	the	the	DET
app01-11088	138	2	final	final	ADJ
app01-11088	138	3	study	study	NOUN
app01-11088	138	4	investigates	investigate	VERB
app01-11088	138	5	how	how	SCONJ
app01-11088	138	6	different	different	ADJ
app01-11088	138	7	area	area	NOUN
app01-11088	138	8	scaling	scale	VERB
app01-11088	138	9	factors	factor	NOUN
app01-11088	138	10	affect	affect	VERB
app01-11088	138	11	the	the	DET
app01-11088	138	12	overall	overall	ADJ
app01-11088	138	13	pixel	pixel	NOUN
app01-11088	138	14	wise	wise	ADJ
app01-11088	138	15	accuracy	accuracy	NOUN
app01-11088	138	16	.	.	PUNCT
app01-11088	139	1	the	the	DET
app01-11088	139	2	results	result	NOUN
app01-11088	139	3	depicted	depict	VERB
app01-11088	139	4	in	in	ADP
app01-11088	139	5	figure	figure	NOUN
app01-11088	139	6	8	8	NUM
app01-11088	139	7	represent	represent	VERB
app01-11088	139	8	the	the	DET
app01-11088	139	9	values	value	NOUN
app01-11088	139	10	of	of	ADP
app01-11088	139	11	the	the	DET
app01-11088	139	12	mean	mean	ADJ
app01-11088	139	13	squared	square	VERB
app01-11088	139	14	error	error	NOUN
app01-11088	139	15	,	,	PUNCT
app01-11088	139	16	measuring	measure	VERB
app01-11088	139	17	the	the	DET
app01-11088	139	18	deviation	deviation	NOUN
app01-11088	139	19	of	of	ADP
app01-11088	139	20	overlap	overlap	NOUN
app01-11088	139	21	between	between	ADP
app01-11088	139	22	the	the	DET
app01-11088	139	23	predicted	predict	VERB
app01-11088	139	24	and	and	CCONJ
app01-11088	139	25	original	original	ADJ
app01-11088	139	26	aggregates	aggregate	NOUN
app01-11088	139	27	.	.	PUNCT
app01-11088	140	1	the	the	DET
app01-11088	140	2	errors	error	NOUN
app01-11088	140	3	have	have	AUX
app01-11088	140	4	been	be	AUX
app01-11088	140	5	adjusted	adjust	VERB
app01-11088	140	6	for	for	ADP
app01-11088	140	7	scaling	scale	VERB
app01-11088	140	8	by	by	ADP
app01-11088	140	9	dividing	divide	VERB
app01-11088	140	10	the	the	DET
app01-11088	140	11	mse	mse	NOUN
app01-11088	140	12	-	-	PUNCT
app01-11088	140	13	error	error	NOUN
app01-11088	140	14	with	with	ADP
app01-11088	140	15	the	the	DET
app01-11088	140	16	square	square	NOUN
app01-11088	140	17	of	of	ADP
app01-11088	140	18	the	the	DET
app01-11088	140	19	scaling	scale	VERB
app01-11088	140	20	factor	factor	NOUN
app01-11088	140	21	.	.	PUNCT
app01-11088	141	1	overall	overall	ADV
app01-11088	141	2	,	,	PUNCT
app01-11088	141	3	the	the	DET
app01-11088	141	4	model	model	NOUN
app01-11088	141	5	demonstrates	demonstrate	VERB
app01-11088	141	6	good	good	ADJ
app01-11088	141	7	fit	fit	NOUN
app01-11088	141	8	between	between	ADP
app01-11088	141	9	the	the	DET
app01-11088	141	10	original	original	ADJ
app01-11088	141	11	aggregates	aggregate	NOUN
app01-11088	141	12	and	and	CCONJ
app01-11088	141	13	the	the	DET
app01-11088	141	14	predicted	predict	VERB
app01-11088	141	15	aggregates	aggregate	NOUN
app01-11088	141	16	for	for	ADP
app01-11088	141	17	a	a	DET
app01-11088	141	18	scaling	scale	VERB
app01-11088	141	19	factor	factor	NOUN
app01-11088	141	20	of	of	ADP
app01-11088	141	21	1	1	NUM
app01-11088	141	22	with	with	ADP
app01-11088	141	23	considerable	considerable	ADJ
app01-11088	141	24	deviation	deviation	NOUN
app01-11088	141	25	from	from	ADP
app01-11088	141	26	the	the	DET
app01-11088	141	27	ideal	ideal	ADJ
app01-11088	141	28	values	value	NOUN
app01-11088	141	29	for	for	ADP
app01-11088	141	30	other	other	ADJ
app01-11088	141	31	scaling	scale	VERB
app01-11088	141	32	factors	factor	NOUN
app01-11088	141	33	.	.	PUNCT
app01-11088	142	1	however	however	ADV
app01-11088	142	2	,	,	PUNCT
app01-11088	142	3	such	such	ADJ
app01-11088	142	4	deviation	deviation	NOUN
app01-11088	142	5	is	be	AUX
app01-11088	142	6	observed	observe	VERB
app01-11088	142	7	to	to	PART
app01-11088	142	8	be	be	AUX
app01-11088	142	9	sample	sample	NOUN
app01-11088	142	10	specific	specific	ADJ
app01-11088	142	11	figure	figure	NOUN
app01-11088	142	12	8	8	NUM
app01-11088	142	13	.	.	PUNCT
app01-11088	142	14	error	error	NOUN
app01-11088	142	15	assessment	assessment	NOUN
app01-11088	142	16	based	base	VERB
app01-11088	142	17	on	on	ADP
app01-11088	142	18	mse	mse	NOUN
app01-11088	142	19	exploring	explore	VERB
app01-11088	142	20	the	the	DET
app01-11088	142	21	overlap	overlap	NOUN
app01-11088	142	22	differences	difference	NOUN
app01-11088	142	23	between	between	ADP
app01-11088	142	24	predicted	predict	VERB
app01-11088	142	25	and	and	CCONJ
app01-11088	142	26	original	original	ADJ
app01-11088	142	27	aggregates	aggregate	NOUN
app01-11088	142	28	.	.	PUNCT
app01-11088	143	1	with	with	SCONJ
app01-11088	143	2	the	the	DET
app01-11088	143	3	deviation	deviation	NOUN
app01-11088	143	4	being	be	AUX
app01-11088	143	5	spread	spread	VERB
app01-11088	143	6	over	over	ADP
app01-11088	143	7	a	a	DET
app01-11088	143	8	wide	wide	ADJ
app01-11088	143	9	range	range	NOUN
app01-11088	143	10	of	of	ADP
app01-11088	143	11	values	value	NOUN
app01-11088	143	12	,	,	PUNCT
app01-11088	143	13	with	with	ADP
app01-11088	143	14	the	the	DET
app01-11088	143	15	model	model	NOUN
app01-11088	143	16	performing	perform	VERB
app01-11088	143	17	well	well	ADV
app01-11088	143	18	in	in	ADP
app01-11088	143	19	some	some	DET
app01-11088	143	20	cases	case	NOUN
app01-11088	143	21	.	.	PUNCT
app01-11088	144	1	nevertheless	nevertheless	ADV
app01-11088	144	2	,	,	PUNCT
app01-11088	144	3	as	as	SCONJ
app01-11088	144	4	has	have	AUX
app01-11088	144	5	already	already	ADV
app01-11088	144	6	been	be	AUX
app01-11088	144	7	highlighted	highlight	VERB
app01-11088	144	8	,	,	PUNCT
app01-11088	144	9	such	such	DET
app01-11088	144	10	a	a	DET
app01-11088	144	11	result	result	NOUN
app01-11088	144	12	is	be	AUX
app01-11088	144	13	expected	expect	VERB
app01-11088	144	14	since	since	SCONJ
app01-11088	144	15	the	the	DET
app01-11088	144	16	model	model	NOUN
app01-11088	144	17	was	be	AUX
app01-11088	144	18	not	not	PART
app01-11088	144	19	explicitly	explicitly	ADV
app01-11088	144	20	trained	train	VERB
app01-11088	144	21	to	to	PART
app01-11088	144	22	factor	factor	VERB
app01-11088	144	23	for	for	ADP
app01-11088	144	24	scaling	scaling	NOUN
app01-11088	144	25	.	.	PUNCT
app01-11088	145	1	4	4	X
app01-11088	145	2	.	.	X
app01-11088	145	3	conclusion	conclusion	NOUN
app01-11088	145	4	this	this	DET
app01-11088	145	5	paper	paper	NOUN
app01-11088	145	6	presents	present	VERB
app01-11088	145	7	a	a	DET
app01-11088	145	8	method	method	NOUN
app01-11088	145	9	for	for	ADP
app01-11088	145	10	generating	generate	VERB
app01-11088	145	11	aggregate	aggregate	ADJ
app01-11088	145	12	shapes	shape	NOUN
app01-11088	145	13	in	in	ADP
app01-11088	145	14	2d	2d	NOUN
app01-11088	145	15	using	use	VERB
app01-11088	145	16	a	a	DET
app01-11088	145	17	deep	deep	ADJ
app01-11088	145	18	-	-	PUNCT
app01-11088	145	19	learning	learn	VERB
app01-11088	145	20	tool	tool	NOUN
app01-11088	145	21	and	and	CCONJ
app01-11088	145	22	a	a	DET
app01-11088	145	23	small	small	ADJ
app01-11088	145	24	set	set	NOUN
app01-11088	145	25	of	of	ADP
app01-11088	145	26	tractable	tractable	ADJ
app01-11088	145	27	and	and	CCONJ
app01-11088	145	28	physically	physically	ADV
app01-11088	145	29	meaningful	meaningful	ADJ
app01-11088	145	30	parameters	parameter	NOUN
app01-11088	145	31	derived	derive	VERB
app01-11088	145	32	from	from	ADP
app01-11088	145	33	moment	moment	NOUN
app01-11088	145	34	invariant	invariant	ADJ
app01-11088	145	35	analysis	analysis	NOUN
app01-11088	145	36	.	.	PUNCT
app01-11088	146	1	this	this	DET
app01-11088	146	2	approach	approach	NOUN
app01-11088	146	3	acts	act	VERB
app01-11088	146	4	as	as	ADP
app01-11088	146	5	the	the	DET
app01-11088	146	6	initial	initial	ADJ
app01-11088	146	7	step	step	NOUN
app01-11088	146	8	in	in	ADP
app01-11088	146	9	developing	develop	VERB
app01-11088	146	10	a	a	DET
app01-11088	146	11	microstructure	microstructure	ADJ
app01-11088	146	12	generator	generator	NOUN
app01-11088	146	13	of	of	ADP
app01-11088	146	14	heterogeneous	heterogeneous	ADJ
app01-11088	146	15	materials	material	NOUN
app01-11088	146	16	that	that	PRON
app01-11088	146	17	relies	rely	VERB
app01-11088	146	18	solely	solely	ADV
app01-11088	146	19	on	on	ADP
app01-11088	146	20	the	the	DET
app01-11088	146	21	inputs	input	NOUN
app01-11088	146	22	which	which	PRON
app01-11088	146	23	have	have	VERB
app01-11088	146	24	clear	clear	ADJ
app01-11088	146	25	and	and	CCONJ
app01-11088	146	26	interpretable	interpretable	ADJ
app01-11088	146	27	meanings	meaning	NOUN
app01-11088	146	28	describing	describe	VERB
app01-11088	146	29	material	material	NOUN
app01-11088	146	30	morphology	morphology	NOUN
app01-11088	146	31	.	.	PUNCT
app01-11088	147	1	the	the	DET
app01-11088	147	2	method	method	NOUN
app01-11088	147	3	demonstrates	demonstrate	VERB
app01-11088	147	4	the	the	DET
app01-11088	147	5	ability	ability	NOUN
app01-11088	147	6	of	of	ADP
app01-11088	147	7	a	a	DET
app01-11088	147	8	neural	neural	ADJ
app01-11088	147	9	-	-	PUNCT
app01-11088	147	10	network	network	NOUN
app01-11088	147	11	model	model	NOUN
app01-11088	147	12	to	to	PART
app01-11088	147	13	recognize	recognize	VERB
app01-11088	147	14	scale	scale	NOUN
app01-11088	147	15	variations	variation	NOUN
app01-11088	147	16	in	in	ADP
app01-11088	147	17	input	input	NOUN
app01-11088	147	18	prompts	prompt	NOUN
app01-11088	147	19	and	and	CCONJ
app01-11088	147	20	adjust	adjust	VERB
app01-11088	147	21	output	output	NOUN
app01-11088	147	22	sizes	size	NOUN
app01-11088	147	23	accordingly	accordingly	ADV
app01-11088	147	24	,	,	PUNCT
app01-11088	147	25	despite	despite	SCONJ
app01-11088	147	26	the	the	DET
app01-11088	147	27	absence	absence	NOUN
app01-11088	147	28	of	of	ADP
app01-11088	147	29	explicit	explicit	ADJ
app01-11088	147	30	training	training	NOUN
app01-11088	147	31	for	for	ADP
app01-11088	147	32	scale	scale	NOUN
app01-11088	147	33	invariance	invariance	NOUN
app01-11088	147	34	.	.	PUNCT
app01-11088	148	1	explicit	explicit	ADJ
app01-11088	148	2	training	training	NOUN
app01-11088	148	3	for	for	ADP
app01-11088	148	4	this	this	DET
app01-11088	148	5	aspect	aspect	NOUN
app01-11088	148	6	is	be	AUX
app01-11088	148	7	expected	expect	VERB
app01-11088	148	8	to	to	PART
app01-11088	148	9	enhance	enhance	VERB
app01-11088	148	10	the	the	DET
app01-11088	148	11	results	result	NOUN
app01-11088	148	12	further	far	ADV
app01-11088	148	13	.	.	PUNCT
app01-11088	149	1	this	this	DET
app01-11088	149	2	method	method	NOUN
app01-11088	149	3	is	be	AUX
app01-11088	149	4	designed	design	VERB
app01-11088	149	5	with	with	ADP
app01-11088	149	6	the	the	DET
app01-11088	149	7	objective	objective	NOUN
app01-11088	149	8	of	of	ADP
app01-11088	149	9	being	be	AUX
app01-11088	149	10	integrating	integrate	VERB
app01-11088	149	11	with	with	ADP
app01-11088	149	12	aggregate	aggregate	ADJ
app01-11088	149	13	packing	packing	NOUN
app01-11088	149	14	schemes	scheme	NOUN
app01-11088	149	15	for	for	ADP
app01-11088	149	16	microstructure	microstructure	ADJ
app01-11088	149	17	generation	generation	NOUN
app01-11088	149	18	of	of	ADP
app01-11088	149	19	concretes	concrete	NOUN
app01-11088	149	20	.	.	PUNCT
app01-11088	150	1	while	while	SCONJ
app01-11088	150	2	the	the	DET
app01-11088	150	3	current	current	ADJ
app01-11088	150	4	results	result	NOUN
app01-11088	150	5	are	be	AUX
app01-11088	150	6	focused	focus	VERB
app01-11088	150	7	on	on	ADP
app01-11088	150	8	2d	2d	NUM
app01-11088	150	9	structures	structure	NOUN
app01-11088	150	10	,	,	PUNCT
app01-11088	150	11	the	the	DET
app01-11088	150	12	concept	concept	NOUN
app01-11088	150	13	can	can	AUX
app01-11088	150	14	be	be	AUX
app01-11088	150	15	easily	easily	ADV
app01-11088	150	16	extended	extend	VERB
app01-11088	150	17	to	to	ADP
app01-11088	150	18	3d	3d	PROPN
app01-11088	150	19	problems	problem	NOUN
app01-11088	150	20	,	,	PUNCT
app01-11088	150	21	offering	offer	VERB
app01-11088	150	22	broader	broad	ADJ
app01-11088	150	23	applicability	applicability	NOUN
app01-11088	150	24	in	in	ADP
app01-11088	150	25	the	the	DET
app01-11088	150	26	study	study	NOUN
app01-11088	150	27	and	and	CCONJ
app01-11088	150	28	design	design	NOUN
app01-11088	150	29	of	of	ADP
app01-11088	150	30	concrete	concrete	ADJ
app01-11088	150	31	microstructures	microstructure	NOUN
app01-11088	150	32	.	.	PUNCT
app01-11088	151	1	acknowledgements	acknowledgement	NOUN
app01-11088	151	2	the	the	DET
app01-11088	151	3	authors	author	NOUN
app01-11088	151	4	are	be	AUX
app01-11088	151	5	thankful	thankful	ADJ
app01-11088	151	6	for	for	ADP
app01-11088	151	7	financial	financial	ADJ
app01-11088	151	8	support	support	NOUN
app01-11088	151	9	from	from	ADP
app01-11088	151	10	the	the	DET
app01-11088	151	11	czech	czech	PROPN
app01-11088	151	12	science	science	NOUN
app01-11088	151	13	foundation	foundation	PROPN
app01-11088	151	14	,	,	PUNCT
app01-11088	151	15	project	project	VERB
app01-11088	151	16	no	no	NOUN
app01-11088	151	17	.	.	NOUN
app01-11088	152	1	22	22	NUM
app01-11088	152	2	-	-	SYM
app01-11088	152	3	35755k	35755k	NUM
app01-11088	152	4	(	(	PUNCT
app01-11088	152	5	kd	kd	PROPN
app01-11088	152	6	,	,	PUNCT
app01-11088	152	7	ak	ak	PROPN
app01-11088	152	8	)	)	PUNCT
app01-11088	152	9	and	and	CCONJ
app01-11088	152	10	the	the	DET
app01-11088	152	11	student	student	NOUN
app01-11088	152	12	grant	grant	NOUN
app01-11088	152	13	competition	competition	NOUN
app01-11088	152	14	of	of	ADP
app01-11088	152	15	ctu	ctu	NOUN
app01-11088	152	16	,	,	PUNCT
app01-11088	152	17	project	project	NOUN
app01-11088	152	18	no	no	NOUN
app01-11088	152	19	.	.	PUNCT
app01-11088	153	1	sgs23/	sgs23/	PROPN
app01-11088	153	2	152	152	NUM
app01-11088	153	3	/	/	SYM
app01-11088	153	4	ohk1/3t/11	ohk1/3t/11	PROPN
app01-11088	153	5	(	(	PUNCT
app01-11088	153	6	js	js	PROPN
app01-11088	153	7	)	)	PUNCT
app01-11088	153	8	.	.	PUNCT
app01-11088	154	1	references	reference	NOUN
app01-11088	154	2	[	[	X
app01-11088	154	3	1	1	NUM
app01-11088	154	4	]	]	X
app01-11088	154	5	h.	h.	PROPN
app01-11088	154	6	c.	c.	PROPN
app01-11088	154	7	sleiman	sleiman	PROPN
app01-11088	154	8	,	,	PUNCT
app01-11088	154	9	m.	m.	PROPN
app01-11088	154	10	h.	h.	PROPN
app01-11088	154	11	moreira	moreira	PROPN
app01-11088	154	12	,	,	PUNCT
app01-11088	154	13	a.	a.	NOUN
app01-11088	154	14	tengattini	tengattini	PROPN
app01-11088	154	15	,	,	PUNCT
app01-11088	154	16	s.	s.	PROPN
app01-11088	154	17	dal	dal	PROPN
app01-11088	154	18	pont	pont	PROPN
app01-11088	154	19	.	.	PUNCT
app01-11088	155	1	from	from	ADP
app01-11088	155	2	tomographic	tomographic	ADJ
app01-11088	155	3	imaging	imaging	NOUN
app01-11088	155	4	to	to	ADP
app01-11088	155	5	numerical	numerical	ADJ
app01-11088	155	6	simulations	simulation	NOUN
app01-11088	155	7	:	:	PUNCT
app01-11088	155	8	an	an	DET
app01-11088	155	9	open	open	ADJ
app01-11088	155	10	-	-	PUNCT
app01-11088	155	11	source	source	NOUN
app01-11088	155	12	workflow	workflow	NOUN
app01-11088	155	13	for	for	ADP
app01-11088	155	14	true	true	ADJ
app01-11088	155	15	21	21	NUM
app01-11088	155	16	k.	k.	PROPN
app01-11088	155	17	das	das	PROPN
app01-11088	155	18	,	,	PUNCT
app01-11088	155	19	j.	j.	PROPN
app01-11088	155	20	sýkora	sýkora	PROPN
app01-11088	155	21	,	,	PUNCT
app01-11088	155	22	a.	a.	PROPN
app01-11088	155	23	kučerová	kučerová	AUX
app01-11088	155	24	acta	acta	PROPN
app01-11088	155	25	polytechnica	polytechnica	PROPN
app01-11088	155	26	ctu	ctu	PROPN
app01-11088	155	27	proceedings	proceeding	NOUN
app01-11088	155	28	morphology	morphology	VERB
app01-11088	155	29	mesoscale	mesoscale	PROPN
app01-11088	155	30	fe	fe	PROPN
app01-11088	155	31	meshes	mesh	NOUN
app01-11088	155	32	.	.	PUNCT
app01-11088	156	1	rilem	rilem	NOUN
app01-11088	156	2	technical	technical	ADJ
app01-11088	156	3	letters	letter	NOUN
app01-11088	156	4	8:158–164	8:158–164	NOUN
app01-11088	156	5	,	,	PUNCT
app01-11088	156	6	2023	2023	NUM
app01-11088	156	7	.	.	PUNCT
app01-11088	157	1	https://doi.org/10.21809/rilemtechlett.2023.184	https://doi.org/10.21809/rilemtechlett.2023.184	VERB
app01-11088	157	2	[	[	X
app01-11088	157	3	2	2	NUM
app01-11088	157	4	]	]	PUNCT
app01-11088	157	5	q.	q.	PROPN
app01-11088	157	6	ren	ren	PROPN
app01-11088	157	7	,	,	PUNCT
app01-11088	157	8	j.	j.	PROPN
app01-11088	157	9	pacheco	pacheco	PROPN
app01-11088	157	10	,	,	PUNCT
app01-11088	157	11	j.	j.	PROPN
app01-11088	157	12	de	de	PROPN
app01-11088	157	13	brito	brito	PROPN
app01-11088	157	14	.	.	PUNCT
app01-11088	158	1	methods	method	NOUN
app01-11088	158	2	for	for	ADP
app01-11088	158	3	the	the	DET
app01-11088	158	4	modelling	modelling	NOUN
app01-11088	158	5	of	of	ADP
app01-11088	158	6	concrete	concrete	ADJ
app01-11088	158	7	mesostructures	mesostructure	NOUN
app01-11088	158	8	:	:	PUNCT
app01-11088	158	9	a	a	DET
app01-11088	158	10	critical	critical	ADJ
app01-11088	158	11	review	review	NOUN
app01-11088	158	12	.	.	PUNCT
app01-11088	159	1	construction	construction	NOUN
app01-11088	159	2	and	and	CCONJ
app01-11088	159	3	building	building	NOUN
app01-11088	159	4	materials	material	NOUN
app01-11088	159	5	408:133570	408:133570	NOUN
app01-11088	159	6	,	,	PUNCT
app01-11088	159	7	2023	2023	NUM
app01-11088	159	8	.	.	PUNCT
app01-11088	160	1	https	https	NOUN
app01-11088	160	2	:	:	PUNCT
app01-11088	160	3	//doi.org/10.1016	//doi.org/10.1016	PROPN
app01-11088	160	4	/	/	SYM
app01-11088	160	5	j.conbuildmat.2023.133570	j.conbuildmat.2023.133570	NOUN
app01-11088	161	1	[	[	X
app01-11088	161	2	3	3	NUM
app01-11088	161	3	]	]	PUNCT
app01-11088	161	4	z.	z.	PROPN
app01-11088	161	5	kammouna	kammouna	PROPN
app01-11088	161	6	,	,	PUNCT
app01-11088	161	7	m.	m.	NOUN
app01-11088	161	8	briffaut	briffaut	PROPN
app01-11088	161	9	,	,	PUNCT
app01-11088	161	10	y.	y.	PROPN
app01-11088	161	11	malecot	malecot	PROPN
app01-11088	161	12	.	.	PUNCT
app01-11088	162	1	mesoscopic	mesoscopic	NOUN
app01-11088	162	2	simulations	simulation	NOUN
app01-11088	162	3	of	of	ADP
app01-11088	162	4	concrete	concrete	ADJ
app01-11088	162	5	strains	strain	NOUN
app01-11088	162	6	incompatibilities	incompatibility	NOUN
app01-11088	162	7	under	under	ADP
app01-11088	162	8	high	high	ADJ
app01-11088	162	9	creep	creep	NOUN
app01-11088	162	10	stress	stress	NOUN
app01-11088	162	11	level	level	NOUN
app01-11088	162	12	and	and	CCONJ
app01-11088	162	13	consequences	consequence	NOUN
app01-11088	162	14	on	on	ADP
app01-11088	162	15	the	the	DET
app01-11088	162	16	mechanical	mechanical	ADJ
app01-11088	162	17	properties	property	NOUN
app01-11088	162	18	.	.	PUNCT
app01-11088	163	1	european	european	ADJ
app01-11088	163	2	journal	journal	PROPN
app01-11088	163	3	of	of	ADP
app01-11088	163	4	environmental	environmental	ADJ
app01-11088	163	5	and	and	CCONJ
app01-11088	163	6	civil	civil	ADJ
app01-11088	163	7	engineering	engineering	NOUN
app01-11088	163	8	23(7):879–893	23(7):879–893	NOUN
app01-11088	163	9	,	,	PUNCT
app01-11088	163	10	2019	2019	NUM
app01-11088	163	11	.	.	PUNCT
app01-11088	164	1	https://doi.org/10.1080/19648189.2017.1320235	https://doi.org/10.1080/19648189.2017.1320235	X
app01-11088	165	1	[	[	X
app01-11088	165	2	4	4	NUM
app01-11088	165	3	]	]	PUNCT
app01-11088	165	4	l.	l.	PROPN
app01-11088	165	5	li	li	PROPN
app01-11088	165	6	,	,	PUNCT
app01-11088	165	7	y.	y.	PROPN
app01-11088	165	8	jin	jin	PROPN
app01-11088	165	9	,	,	PUNCT
app01-11088	165	10	y.	y.	PROPN
app01-11088	165	11	jia	jia	PROPN
app01-11088	165	12	,	,	PUNCT
app01-11088	165	13	et	et	PROPN
app01-11088	165	14	al	al	PROPN
app01-11088	165	15	.	.	PUNCT
app01-11088	165	16	influence	influence	NOUN
app01-11088	165	17	of	of	ADP
app01-11088	165	18	inclusion	inclusion	NOUN
app01-11088	165	19	rigidity	rigidity	NOUN
app01-11088	165	20	on	on	ADP
app01-11088	165	21	shrinkage	shrinkage	NOUN
app01-11088	165	22	induced	induce	VERB
app01-11088	165	23	micro	micro	NOUN
app01-11088	165	24	-	-	NOUN
app01-11088	165	25	cracking	cracking	NOUN
app01-11088	165	26	of	of	ADP
app01-11088	165	27	cementitious	cementitious	ADJ
app01-11088	165	28	materials	material	NOUN
app01-11088	165	29	.	.	PUNCT
app01-11088	166	1	cement	cement	NOUN
app01-11088	166	2	and	and	CCONJ
app01-11088	166	3	concrete	concrete	ADJ
app01-11088	166	4	composites	composite	NOUN
app01-11088	166	5	114:103773	114:103773	NUM
app01-11088	166	6	,	,	PUNCT
app01-11088	166	7	2020	2020	NUM
app01-11088	166	8	.	.	PUNCT
app01-11088	167	1	https	https	NOUN
app01-11088	167	2	:	:	PUNCT
app01-11088	167	3	//doi.org/10.1016	//doi.org/10.1016	PROPN
app01-11088	167	4	/	/	SYM
app01-11088	167	5	j.cemconcomp.2020.103773	j.cemconcomp.2020.103773	PROPN
app01-11088	168	1	[	[	X
app01-11088	168	2	5	5	X
app01-11088	168	3	]	]	PUNCT
app01-11088	168	4	z.	z.	PROPN
app01-11088	168	5	zheng	zheng	PROPN
app01-11088	168	6	,	,	PUNCT
app01-11088	168	7	x.	x.	PROPN
app01-11088	168	8	wei	wei	PROPN
app01-11088	168	9	.	.	PROPN
app01-11088	168	10	mesoscopic	mesoscopic	NOUN
app01-11088	168	11	models	model	NOUN
app01-11088	168	12	and	and	CCONJ
app01-11088	168	13	numerical	numerical	ADJ
app01-11088	168	14	simulations	simulation	NOUN
app01-11088	168	15	of	of	ADP
app01-11088	168	16	the	the	DET
app01-11088	168	17	temperature	temperature	NOUN
app01-11088	168	18	field	field	NOUN
app01-11088	168	19	and	and	CCONJ
app01-11088	168	20	hydration	hydration	NOUN
app01-11088	168	21	degree	degree	NOUN
app01-11088	168	22	in	in	ADP
app01-11088	168	23	early	early	ADJ
app01-11088	168	24	-	-	PUNCT
app01-11088	168	25	age	age	NOUN
app01-11088	168	26	concrete	concrete	NOUN
app01-11088	168	27	.	.	PUNCT
app01-11088	169	1	construction	construction	NOUN
app01-11088	169	2	and	and	CCONJ
app01-11088	169	3	building	building	NOUN
app01-11088	169	4	materials	material	NOUN
app01-11088	169	5	266:121001	266:121001	NUM
app01-11088	169	6	,	,	PUNCT
app01-11088	169	7	2021	2021	NUM
app01-11088	169	8	.	.	PUNCT
app01-11088	170	1	https	https	NOUN
app01-11088	170	2	:	:	PUNCT
app01-11088	170	3	//doi.org/10.1016	//doi.org/10.1016	PROPN
app01-11088	170	4	/	/	SYM
app01-11088	170	5	j.conbuildmat.2020.121001	j.conbuildmat.2020.121001	NOUN
app01-11088	170	6	[	[	X
app01-11088	170	7	6	6	NUM
app01-11088	170	8	]	]	PUNCT
app01-11088	170	9	t.-d	t.-d	NOUN
app01-11088	170	10	.	.	PUNCT
app01-11088	171	1	nguyen	nguyen	NOUN
app01-11088	171	2	,	,	PUNCT
app01-11088	171	3	d.-t	d.-t	NOUN
app01-11088	171	4	.	.	PUNCT
app01-11088	172	1	pham	pham	PROPN
app01-11088	172	2	,	,	PUNCT
app01-11088	172	3	m.-n	m.-n	PROPN
app01-11088	172	4	.	.	PUNCT
app01-11088	173	1	vu	vu	PROPN
app01-11088	173	2	.	.	PUNCT
app01-11088	173	3	thermomechanically	thermomechanically	ADV
app01-11088	173	4	-	-	PUNCT
app01-11088	173	5	induced	induce	VERB
app01-11088	173	6	thermal	thermal	ADJ
app01-11088	173	7	conductivity	conductivity	NOUN
app01-11088	173	8	change	change	NOUN
app01-11088	173	9	and	and	CCONJ
app01-11088	173	10	its	its	PRON
app01-11088	173	11	effect	effect	NOUN
app01-11088	173	12	on	on	ADP
app01-11088	173	13	the	the	DET
app01-11088	173	14	behaviour	behaviour	NOUN
app01-11088	173	15	of	of	ADP
app01-11088	173	16	concrete	concrete	NOUN
app01-11088	173	17	.	.	PUNCT
app01-11088	174	1	construction	construction	NOUN
app01-11088	174	2	and	and	CCONJ
app01-11088	174	3	building	building	NOUN
app01-11088	174	4	materials	material	NOUN
app01-11088	174	5	198:98–105	198:98–105	NUM
app01-11088	174	6	,	,	PUNCT
app01-11088	174	7	2019	2019	NUM
app01-11088	174	8	.	.	PUNCT
app01-11088	175	1	https	https	NOUN
app01-11088	175	2	:	:	PUNCT
app01-11088	176	1	//doi.org/10.1016	//doi.org/10.1016	PROPN
app01-11088	176	2	/	/	SYM
app01-11088	176	3	j.conbuildmat.2018.11.146	j.conbuildmat.2018.11.146	PROPN
app01-11088	177	1	[	[	X
app01-11088	177	2	7	7	NUM
app01-11088	177	3	]	]	PUNCT
app01-11088	177	4	a.	a.	NOUN
app01-11088	177	5	thirumalaiselvi	thirumalaiselvi	NOUN
app01-11088	177	6	,	,	PUNCT
app01-11088	177	7	n.	n.	PROPN
app01-11088	177	8	anandavalli	anandavalli	PROPN
app01-11088	177	9	,	,	PUNCT
app01-11088	177	10	j.	j.	PROPN
app01-11088	177	11	rajasankar	rajasankar	PROPN
app01-11088	177	12	.	.	PUNCT
app01-11088	178	1	mesoscale	mesoscale	ADJ
app01-11088	178	2	studies	study	NOUN
app01-11088	178	3	on	on	ADP
app01-11088	178	4	the	the	DET
app01-11088	178	5	effect	effect	NOUN
app01-11088	178	6	of	of	ADP
app01-11088	178	7	aggregate	aggregate	ADJ
app01-11088	178	8	shape	shape	NOUN
app01-11088	178	9	idealisation	idealisation	NOUN
app01-11088	178	10	in	in	ADP
app01-11088	178	11	concrete	concrete	NOUN
app01-11088	178	12	.	.	PUNCT
app01-11088	179	1	magazine	magazine	NOUN
app01-11088	179	2	of	of	ADP
app01-11088	179	3	concrete	concrete	ADJ
app01-11088	179	4	research	research	NOUN
app01-11088	179	5	71(5):244–259	71(5):244–259	PROPN
app01-11088	179	6	,	,	PUNCT
app01-11088	179	7	2019	2019	NUM
app01-11088	179	8	.	.	PUNCT
app01-11088	180	1	https://doi.org/10.1680/jmacr.17.00184	https://doi.org/10.1680/jmacr.17.00184	X
app01-11088	180	2	[	[	X
app01-11088	180	3	8	8	NUM
app01-11088	180	4	]	]	PUNCT
app01-11088	180	5	k.	k.	PROPN
app01-11088	180	6	l.	l.	PROPN
app01-11088	180	7	scrivener	scrivener	PROPN
app01-11088	180	8	,	,	PUNCT
app01-11088	180	9	a.	a.	PROPN
app01-11088	180	10	k.	k.	PROPN
app01-11088	180	11	crumbie	crumbie	PROPN
app01-11088	180	12	,	,	PUNCT
app01-11088	180	13	p.	p.	PROPN
app01-11088	180	14	laugesen	laugesen	PROPN
app01-11088	180	15	.	.	PUNCT
app01-11088	181	1	the	the	DET
app01-11088	181	2	interfacial	interfacial	ADJ
app01-11088	181	3	transition	transition	NOUN
app01-11088	181	4	zone	zone	NOUN
app01-11088	181	5	(	(	PUNCT
app01-11088	181	6	itz	itz	PROPN
app01-11088	181	7	)	)	PUNCT
app01-11088	181	8	between	between	ADP
app01-11088	181	9	cement	cement	NOUN
app01-11088	181	10	paste	paste	NOUN
app01-11088	181	11	and	and	CCONJ
app01-11088	181	12	aggregate	aggregate	VERB
app01-11088	181	13	in	in	ADP
app01-11088	181	14	concrete	concrete	NOUN
app01-11088	181	15	.	.	PUNCT
app01-11088	182	1	interface	interface	NOUN
app01-11088	182	2	science	science	NOUN
app01-11088	182	3	12:411–421	12:411–421	PROPN
app01-11088	182	4	,	,	PUNCT
app01-11088	182	5	2004	2004	NUM
app01-11088	182	6	.	.	PUNCT
app01-11088	183	1	https://doi.org/10.1023/b	https://doi.org/10.1023/b	NOUN
app01-11088	183	2	:	:	PUNCT
app01-11088	183	3	ints.0000042339.92990.4c	ints.0000042339.92990.4c	PROPN
app01-11088	184	1	[	[	X
app01-11088	184	2	9	9	NUM
app01-11088	184	3	]	]	PUNCT
app01-11088	184	4	m.	m.	NOUN
app01-11088	184	5	nitka	nitka	PROPN
app01-11088	184	6	,	,	PUNCT
app01-11088	184	7	j.	j.	PROPN
app01-11088	184	8	tejchman	tejchman	PROPN
app01-11088	184	9	.	.	PUNCT
app01-11088	185	1	meso	meso	NOUN
app01-11088	185	2	-	-	PUNCT
app01-11088	185	3	mechanical	mechanical	ADJ
app01-11088	185	4	modelling	modelling	NOUN
app01-11088	185	5	of	of	ADP
app01-11088	185	6	damage	damage	NOUN
app01-11088	185	7	in	in	ADP
app01-11088	185	8	concrete	concrete	NOUN
app01-11088	185	9	using	use	VERB
app01-11088	185	10	discrete	discrete	ADJ
app01-11088	185	11	element	element	NOUN
app01-11088	185	12	method	method	NOUN
app01-11088	185	13	with	with	ADP
app01-11088	185	14	porous	porous	ADJ
app01-11088	185	15	itzs	itz	NOUN
app01-11088	185	16	of	of	ADP
app01-11088	185	17	defined	define	VERB
app01-11088	185	18	width	width	NOUN
app01-11088	185	19	around	around	ADP
app01-11088	185	20	aggregates	aggregate	NOUN
app01-11088	185	21	.	.	PUNCT
app01-11088	186	1	engineering	engineering	NOUN
app01-11088	186	2	fracture	fracture	PROPN
app01-11088	186	3	mechanics	mechanic	NOUN
app01-11088	186	4	231:107029	231:107029	NUM
app01-11088	186	5	,	,	PUNCT
app01-11088	186	6	2020	2020	NUM
app01-11088	186	7	.	.	PUNCT
app01-11088	187	1	https	https	NOUN
app01-11088	187	2	:	:	PUNCT
app01-11088	187	3	//doi.org/10.1016	//doi.org/10.1016	PROPN
app01-11088	187	4	/	/	SYM
app01-11088	187	5	j.engfracmech.2020.107029	j.engfracmech.2020.107029	NOUN
app01-11088	188	1	[	[	X
app01-11088	188	2	10	10	NUM
app01-11088	188	3	]	]	PUNCT
app01-11088	188	4	k.	k.	PROPN
app01-11088	188	5	matouš	matouš	PROPN
app01-11088	188	6	,	,	PUNCT
app01-11088	188	7	m.	m.	NOUN
app01-11088	188	8	g.	g.	PROPN
app01-11088	188	9	geers	geers	PROPN
app01-11088	188	10	,	,	PUNCT
app01-11088	188	11	v.	v.	PROPN
app01-11088	188	12	g.	g.	PROPN
app01-11088	188	13	kouznetsova	kouznetsova	PROPN
app01-11088	188	14	,	,	PUNCT
app01-11088	188	15	a.	a.	PROPN
app01-11088	188	16	gillman	gillman	PROPN
app01-11088	188	17	.	.	PUNCT
app01-11088	189	1	a	a	DET
app01-11088	189	2	review	review	NOUN
app01-11088	189	3	of	of	ADP
app01-11088	189	4	predictive	predictive	ADJ
app01-11088	189	5	nonlinear	nonlinear	ADJ
app01-11088	189	6	theories	theory	NOUN
app01-11088	189	7	for	for	ADP
app01-11088	189	8	multiscale	multiscale	ADJ
app01-11088	189	9	modeling	modeling	NOUN
app01-11088	189	10	of	of	ADP
app01-11088	189	11	heterogeneous	heterogeneous	ADJ
app01-11088	189	12	materials	material	NOUN
app01-11088	189	13	.	.	PUNCT
app01-11088	190	1	journal	journal	NOUN
app01-11088	190	2	of	of	ADP
app01-11088	190	3	computational	computational	ADJ
app01-11088	190	4	physics	physic	NOUN
app01-11088	190	5	330:192–220	330:192–220	NUM
app01-11088	190	6	,	,	PUNCT
app01-11088	190	7	2017	2017	NUM
app01-11088	190	8	.	.	PUNCT
app01-11088	191	1	https://doi.org/10.1016/j.jcp.2016.10.070	https://doi.org/10.1016/j.jcp.2016.10.070	ADP
app01-11088	192	1	[	[	X
app01-11088	192	2	11	11	NUM
app01-11088	192	3	]	]	X
app01-11088	192	4	r.	r.	PROPN
app01-11088	192	5	bostanabad	bostanabad	PROPN
app01-11088	192	6	,	,	PUNCT
app01-11088	192	7	y.	y.	PROPN
app01-11088	192	8	zhang	zhang	PROPN
app01-11088	192	9	,	,	PUNCT
app01-11088	192	10	x.	x.	PROPN
app01-11088	192	11	li	li	PROPN
app01-11088	192	12	,	,	PUNCT
app01-11088	192	13	et	et	PROPN
app01-11088	192	14	al	al	PROPN
app01-11088	192	15	.	.	PUNCT
app01-11088	192	16	computational	computational	ADJ
app01-11088	192	17	microstructure	microstructure	ADJ
app01-11088	192	18	characterization	characterization	NOUN
app01-11088	192	19	and	and	CCONJ
app01-11088	192	20	reconstruction	reconstruction	NOUN
app01-11088	192	21	:	:	PUNCT
app01-11088	192	22	review	review	NOUN
app01-11088	192	23	of	of	ADP
app01-11088	192	24	the	the	DET
app01-11088	192	25	state	state	NOUN
app01-11088	192	26	-	-	PUNCT
app01-11088	192	27	of	of	ADP
app01-11088	192	28	-	-	PUNCT
app01-11088	192	29	the	the	DET
app01-11088	192	30	-	-	PUNCT
app01-11088	192	31	art	art	NOUN
app01-11088	192	32	techniques	technique	NOUN
app01-11088	192	33	.	.	PUNCT
app01-11088	193	1	progress	progress	NOUN
app01-11088	193	2	in	in	ADP
app01-11088	193	3	materials	material	NOUN
app01-11088	193	4	science	science	NOUN
app01-11088	193	5	95:1–41	95:1–41	NUM
app01-11088	193	6	,	,	PUNCT
app01-11088	193	7	2018	2018	NUM
app01-11088	193	8	.	.	PUNCT
app01-11088	194	1	https://doi.org/10.1016/j.pmatsci.2018.01.005	https://doi.org/10.1016/j.pmatsci.2018.01.005	NUM
app01-11088	195	1	[	[	X
app01-11088	195	2	12	12	NUM
app01-11088	195	3	]	]	X
app01-11088	195	4	f.-g	f.-g	PROPN
app01-11088	195	5	.	.	PUNCT
app01-11088	196	1	guo	guo	PROPN
app01-11088	196	2	,	,	PUNCT
app01-11088	196	3	h.	h.	PROPN
app01-11088	196	4	zhang	zhang	PROPN
app01-11088	196	5	,	,	PUNCT
app01-11088	196	6	z.-j	z.-j	PROPN
app01-11088	196	7	.	.	PUNCT
app01-11088	197	1	yang	yang	PROPN
app01-11088	197	2	,	,	PUNCT
app01-11088	197	3	et	et	PROPN
app01-11088	197	4	al	al	PROPN
app01-11088	197	5	.	.	PUNCT
app01-11088	198	1	a	a	DET
app01-11088	198	2	spherical	spherical	ADJ
app01-11088	198	3	harmonic	harmonic	ADJ
app01-11088	198	4	-	-	PUNCT
app01-11088	198	5	random	random	ADJ
app01-11088	198	6	field	field	NOUN
app01-11088	198	7	coupled	couple	VERB
app01-11088	198	8	method	method	NOUN
app01-11088	198	9	for	for	ADP
app01-11088	198	10	efficient	efficient	ADJ
app01-11088	198	11	reconstruction	reconstruction	NOUN
app01-11088	198	12	of	of	ADP
app01-11088	198	13	ct	ct	NOUN
app01-11088	198	14	-	-	PUNCT
app01-11088	198	15	image	image	NOUN
app01-11088	198	16	based	base	VERB
app01-11088	198	17	3d	3d	PROPN
app01-11088	198	18	aggregates	aggregate	NOUN
app01-11088	198	19	with	with	ADP
app01-11088	198	20	controllable	controllable	ADJ
app01-11088	198	21	multiscale	multiscale	ADJ
app01-11088	198	22	morphology	morphology	NOUN
app01-11088	198	23	.	.	PUNCT
app01-11088	199	1	computer	computer	NOUN
app01-11088	199	2	methods	method	NOUN
app01-11088	199	3	in	in	ADP
app01-11088	199	4	applied	applied	ADJ
app01-11088	199	5	mechanics	mechanic	NOUN
app01-11088	199	6	and	and	CCONJ
app01-11088	199	7	engineering	engineering	NOUN
app01-11088	199	8	406:115901	406:115901	NUM
app01-11088	199	9	,	,	PUNCT
app01-11088	199	10	2023	2023	NUM
app01-11088	199	11	.	.	PUNCT
app01-11088	199	12	https://doi.org/10.1016/j.cma.2023.115901	https://doi.org/10.1016/j.cma.2023.115901	PROPN
app01-11088	200	1	[	[	X
app01-11088	200	2	13	13	NUM
app01-11088	200	3	]	]	PUNCT
app01-11088	200	4	h.	h.	PROPN
app01-11088	200	5	xu	xu	PROPN
app01-11088	200	6	,	,	PUNCT
app01-11088	200	7	r.	r.	PROPN
app01-11088	200	8	liu	liu	PROPN
app01-11088	200	9	,	,	PUNCT
app01-11088	200	10	a.	a.	PROPN
app01-11088	200	11	choudhary	choudhary	PROPN
app01-11088	200	12	,	,	PUNCT
app01-11088	200	13	w.	w.	PROPN
app01-11088	200	14	chen	chen	PROPN
app01-11088	200	15	.	.	PUNCT
app01-11088	201	1	a	a	DET
app01-11088	201	2	machine	machine	NOUN
app01-11088	201	3	learning	learning	NOUN
app01-11088	201	4	-	-	PUNCT
app01-11088	201	5	based	base	VERB
app01-11088	201	6	design	design	NOUN
app01-11088	201	7	representation	representation	NOUN
app01-11088	201	8	method	method	NOUN
app01-11088	201	9	for	for	ADP
app01-11088	201	10	designing	design	VERB
app01-11088	201	11	heterogeneous	heterogeneous	ADJ
app01-11088	201	12	microstructures	microstructure	NOUN
app01-11088	201	13	.	.	PUNCT
app01-11088	202	1	journal	journal	NOUN
app01-11088	202	2	of	of	ADP
app01-11088	202	3	mechanical	mechanical	ADJ
app01-11088	202	4	design	design	PROPN
app01-11088	202	5	137(5):051403	137(5):051403	NOUN
app01-11088	202	6	,	,	PUNCT
app01-11088	202	7	2015	2015	NUM
app01-11088	202	8	.	.	PUNCT
app01-11088	203	1	https://doi.org/10.1115/1.4029768	https://doi.org/10.1115/1.4029768	NOUN
app01-11088	204	1	[	[	X
app01-11088	204	2	14	14	NUM
app01-11088	204	3	]	]	PUNCT
app01-11088	204	4	m.-k	m.-k	PROPN
app01-11088	204	5	.	.	PUNCT
app01-11088	205	1	hu	hu	PROPN
app01-11088	205	2	.	.	PROPN
app01-11088	206	1	visual	visual	ADJ
app01-11088	206	2	pattern	pattern	NOUN
app01-11088	206	3	recognition	recognition	NOUN
app01-11088	206	4	by	by	ADP
app01-11088	206	5	moment	moment	NOUN
app01-11088	206	6	invariants	invariant	NOUN
app01-11088	206	7	.	.	PUNCT
app01-11088	207	1	ire	ire	ADJ
app01-11088	207	2	transactions	transaction	NOUN
app01-11088	207	3	on	on	ADP
app01-11088	207	4	information	information	NOUN
app01-11088	207	5	theory	theory	NOUN
app01-11088	207	6	8(2):179–187	8(2):179–187	NUM
app01-11088	207	7	,	,	PUNCT
app01-11088	207	8	1962	1962	NUM
app01-11088	207	9	.	.	PUNCT
app01-11088	208	1	https://doi.org/10.1109/tit.1962.1057692	https://doi.org/10.1109/tit.1962.1057692	PROPN
app01-11088	209	1	[	[	X
app01-11088	209	2	15	15	NUM
app01-11088	209	3	]	]	X
app01-11088	209	4	j.	j.	PROPN
app01-11088	209	5	flusser	flusser	PROPN
app01-11088	209	6	,	,	PUNCT
app01-11088	209	7	t.	t.	PROPN
app01-11088	209	8	suk	suk	PROPN
app01-11088	209	9	.	.	PROPN
app01-11088	209	10	pattern	pattern	NOUN
app01-11088	209	11	recognition	recognition	NOUN
app01-11088	209	12	by	by	ADP
app01-11088	209	13	affine	affine	ADJ
app01-11088	209	14	moment	moment	NOUN
app01-11088	209	15	invariants	invariant	NOUN
app01-11088	209	16	.	.	PUNCT
app01-11088	210	1	pattern	pattern	NOUN
app01-11088	210	2	recognition	recognition	NOUN
app01-11088	210	3	26(1):167–174	26(1):167–174	NUM
app01-11088	210	4	,	,	PUNCT
app01-11088	210	5	1993	1993	NUM
app01-11088	210	6	.	.	PUNCT
app01-11088	211	1	https://doi.org/10.1016/0031-3203(93)90098-h	https://doi.org/10.1016/0031-3203(93)90098-h	PROPN
app01-11088	212	1	[	[	X
app01-11088	212	2	16	16	NUM
app01-11088	212	3	]	]	X
app01-11088	212	4	g.	g.	PROPN
app01-11088	212	5	a.	a.	NOUN
app01-11088	212	6	papakostas	papakosta	NOUN
app01-11088	212	7	.	.	PUNCT
app01-11088	213	1	over	over	ADP
app01-11088	213	2	50	50	NUM
app01-11088	213	3	years	year	NOUN
app01-11088	213	4	of	of	ADP
app01-11088	213	5	image	image	NOUN
app01-11088	213	6	moments	moment	NOUN
app01-11088	213	7	and	and	CCONJ
app01-11088	213	8	moment	moment	NOUN
app01-11088	213	9	invariants	invariant	NOUN
app01-11088	213	10	.	.	PUNCT
app01-11088	214	1	gate	gate	NOUN
app01-11088	214	2	to	to	ADP
app01-11088	214	3	computer	computer	NOUN
app01-11088	214	4	science	science	PROPN
app01-11088	214	5	&	&	CCONJ
app01-11088	214	6	research	research	PROPN
app01-11088	214	7	1	1	NUM
app01-11088	214	8	,	,	PUNCT
app01-11088	214	9	2014	2014	NUM
app01-11088	214	10	.	.	PUNCT
app01-11088	215	1	https://doi.org/10.15579/gcsr.vol1.ch1	https://doi.org/10.15579/gcsr.vol1.ch1	PROPN
app01-11088	216	1	[	[	X
app01-11088	216	2	17	17	NUM
app01-11088	216	3	]	]	PUNCT
app01-11088	216	4	m.	m.	PROPN
app01-11088	216	5	d.	d.	PROPN
app01-11088	216	6	zeiler	zeiler	PROPN
app01-11088	216	7	,	,	PUNCT
app01-11088	216	8	r.	r.	PROPN
app01-11088	216	9	fergus	fergus	PROPN
app01-11088	216	10	.	.	PUNCT
app01-11088	217	1	visualizing	visualize	VERB
app01-11088	217	2	and	and	CCONJ
app01-11088	217	3	understanding	understand	VERB
app01-11088	217	4	convolutional	convolutional	ADJ
app01-11088	217	5	networks	network	NOUN
app01-11088	217	6	.	.	PUNCT
app01-11088	218	1	in	in	ADP
app01-11088	218	2	computer	computer	NOUN
app01-11088	218	3	vision	vision	NOUN
app01-11088	218	4	–	–	PUNCT
app01-11088	218	5	eccv	eccv	ADJ
app01-11088	218	6	2014	2014	NUM
app01-11088	218	7	:	:	PUNCT
app01-11088	218	8	13th	13th	ADJ
app01-11088	218	9	european	european	PROPN
app01-11088	218	10	conference	conference	PROPN
app01-11088	218	11	,	,	PUNCT
app01-11088	218	12	zurich	zurich	PROPN
app01-11088	218	13	,	,	PUNCT
app01-11088	218	14	switzerland	switzerland	PROPN
app01-11088	218	15	,	,	PUNCT
app01-11088	218	16	september	september	PROPN
app01-11088	218	17	6	6	NUM
app01-11088	218	18	-	-	SYM
app01-11088	218	19	12	12	NUM
app01-11088	218	20	,	,	PUNCT
app01-11088	218	21	2014	2014	NUM
app01-11088	218	22	,	,	PUNCT
app01-11088	218	23	proceedings	proceeding	NOUN
app01-11088	218	24	,	,	PUNCT
app01-11088	218	25	part	part	NOUN
app01-11088	218	26	i	i	PRON
app01-11088	218	27	13	13	NUM
app01-11088	218	28	,	,	PUNCT
app01-11088	218	29	pp	pp	ADJ
app01-11088	218	30	.	.	PUNCT
app01-11088	219	1	818–833	818–833	NUM
app01-11088	219	2	.	.	PUNCT
app01-11088	219	3	springer	springer	NOUN
app01-11088	219	4	,	,	PUNCT
app01-11088	219	5	2014	2014	NUM
app01-11088	219	6	.	.	PUNCT
app01-11088	220	1	https://doi.org/10.1007/978-3-319-10590-1_53	https://doi.org/10.1007/978-3-319-10590-1_53	PROPN
app01-11088	221	1	[	[	X
app01-11088	221	2	18	18	NUM
app01-11088	221	3	]	]	PUNCT
app01-11088	221	4	a.	a.	NOUN
app01-11088	221	5	paszke	paszke	NOUN
app01-11088	221	6	,	,	PUNCT
app01-11088	221	7	s.	s.	PROPN
app01-11088	221	8	gross	gross	PROPN
app01-11088	221	9	,	,	PUNCT
app01-11088	221	10	f.	f.	PROPN
app01-11088	221	11	massa	massa	PROPN
app01-11088	221	12	,	,	PUNCT
app01-11088	221	13	et	et	PROPN
app01-11088	221	14	al	al	PROPN
app01-11088	221	15	.	.	PROPN
app01-11088	221	16	pytorch	pytorch	PROPN
app01-11088	221	17	:	:	PUNCT
app01-11088	221	18	an	an	DET
app01-11088	221	19	imperative	imperative	ADJ
app01-11088	221	20	style	style	NOUN
app01-11088	221	21	,	,	PUNCT
app01-11088	221	22	high	high	ADJ
app01-11088	221	23	-	-	PUNCT
app01-11088	221	24	performance	performance	NOUN
app01-11088	221	25	deep	deep	ADJ
app01-11088	221	26	learning	learning	NOUN
app01-11088	221	27	library	library	NOUN
app01-11088	221	28	.	.	PUNCT
app01-11088	222	1	in	in	ADP
app01-11088	222	2	advances	advance	NOUN
app01-11088	222	3	in	in	ADP
app01-11088	222	4	neural	neural	ADJ
app01-11088	222	5	information	information	NOUN
app01-11088	222	6	processing	processing	NOUN
app01-11088	222	7	systems	system	NOUN
app01-11088	222	8	,	,	PUNCT
app01-11088	222	9	vol	vol	NOUN
app01-11088	222	10	.	.	PUNCT
app01-11088	222	11	32	32	NUM
app01-11088	222	12	.	.	X
app01-11088	222	13	2019	2019	NUM
app01-11088	222	14	.	.	PUNCT
app01-11088	223	1	[	[	X
app01-11088	223	2	19	19	NUM
app01-11088	223	3	]	]	X
app01-11088	223	4	d.	d.	PROPN
app01-11088	223	5	p.	p.	PROPN
app01-11088	223	6	kingma	kingma	PROPN
app01-11088	223	7	,	,	PUNCT
app01-11088	223	8	j.	j.	PROPN
app01-11088	223	9	ba	ba	PROPN
app01-11088	223	10	.	.	PUNCT
app01-11088	224	1	adam	adam	PROPN
app01-11088	224	2	:	:	PUNCT
app01-11088	224	3	a	a	DET
app01-11088	224	4	method	method	NOUN
app01-11088	224	5	for	for	ADP
app01-11088	224	6	stochastic	stochastic	ADJ
app01-11088	224	7	optimization	optimization	NOUN
app01-11088	224	8	,	,	PUNCT
app01-11088	224	9	2014	2014	NUM
app01-11088	224	10	.	.	PUNCT
app01-11088	225	1	https://doi.org/10.48550/arxiv.1412.6980	https://doi.org/10.48550/arxiv.1412.6980	PROPN
app01-11088	226	1	[	[	X
app01-11088	226	2	20	20	NUM
app01-11088	226	3	]	]	PUNCT
app01-11088	226	4	s.	s.	PROPN
app01-11088	226	5	van	van	PROPN
app01-11088	226	6	der	der	PROPN
app01-11088	226	7	walt	walt	PROPN
app01-11088	226	8	,	,	PUNCT
app01-11088	226	9	j.	j.	PROPN
app01-11088	226	10	l.	l.	PROPN
app01-11088	226	11	schönberger	schönberger	PROPN
app01-11088	226	12	,	,	PUNCT
app01-11088	226	13	j.	j.	PROPN
app01-11088	226	14	nunez	nunez	PROPN
app01-11088	226	15	-	-	PUNCT
app01-11088	226	16	iglesias	iglesias	PROPN
app01-11088	226	17	,	,	PUNCT
app01-11088	226	18	et	et	PROPN
app01-11088	226	19	al	al	PROPN
app01-11088	226	20	.	.	PUNCT
app01-11088	226	21	scikit	scikit	NOUN
app01-11088	226	22	-	-	PUNCT
app01-11088	226	23	image	image	NOUN
app01-11088	226	24	:	:	PUNCT
app01-11088	226	25	image	image	NOUN
app01-11088	226	26	processing	processing	NOUN
app01-11088	226	27	in	in	ADP
app01-11088	226	28	python	python	PROPN
app01-11088	226	29	.	.	PUNCT
app01-11088	227	1	peerj	peerj	PROPN
app01-11088	227	2	2	2	NUM
app01-11088	227	3	:	:	PUNCT
app01-11088	227	4	e453	e453	NUM
app01-11088	227	5	,	,	PUNCT
app01-11088	227	6	2014	2014	NUM
app01-11088	227	7	.	.	PUNCT
app01-11088	228	1	https://doi.org/10.7717/peerj.453	https://doi.org/10.7717/peerj.453	NOUN
app01-11088	228	2	22	22	NUM
app01-11088	228	3	https://doi.org/10.21809/rilemtechlett.2023.184	https://doi.org/10.21809/rilemtechlett.2023.184	NOUN
app01-11088	228	4	https://doi.org/10.1016/j.conbuildmat.2023.133570	https://doi.org/10.1016/j.conbuildmat.2023.133570	NOUN
app01-11088	228	5	https://doi.org/10.1016/j.conbuildmat.2023.133570	https://doi.org/10.1016/j.conbuildmat.2023.133570	NOUN
app01-11088	228	6	https://doi.org/10.1080/19648189.2017.1320235	https://doi.org/10.1080/19648189.2017.1320235	PROPN
app01-11088	229	1	https://doi.org/10.1016/j.cemconcomp.2020.103773	https://doi.org/10.1016/j.cemconcomp.2020.103773	ADJ
app01-11088	229	2	https://doi.org/10.1016/j.cemconcomp.2020.103773	https://doi.org/10.1016/j.cemconcomp.2020.103773	PROPN
app01-11088	229	3	https://doi.org/10.1016/j.conbuildmat.2020.121001	https://doi.org/10.1016/j.conbuildmat.2020.121001	PROPN
app01-11088	229	4	https://doi.org/10.1016/j.conbuildmat.2020.121001	https://doi.org/10.1016/j.conbuildmat.2020.121001	NOUN
app01-11088	229	5	https://doi.org/10.1016/j.conbuildmat.2018.11.146	https://doi.org/10.1016/j.conbuildmat.2018.11.146	NOUN
app01-11088	229	6	https://doi.org/10.1016/j.conbuildmat.2018.11.146	https://doi.org/10.1016/j.conbuildmat.2018.11.146	PROPN
app01-11088	229	7	https://doi.org/10.1680/jmacr.17.00184	https://doi.org/10.1680/jmacr.17.00184	NOUN
app01-11088	229	8	https://doi.org/10.1023/b:ints.0000042339.92990.4c	https://doi.org/10.1023/b:ints.0000042339.92990.4c	PROPN
app01-11088	229	9	https://doi.org/10.1023/b:ints.0000042339.92990.4c	https://doi.org/10.1023/b:ints.0000042339.92990.4c	PROPN
app01-11088	229	10	https://doi.org/10.1016/j.engfracmech.2020.107029	https://doi.org/10.1016/j.engfracmech.2020.107029	PROPN
app01-11088	229	11	https://doi.org/10.1016/j.engfracmech.2020.107029	https://doi.org/10.1016/j.engfracmech.2020.107029	PROPN
app01-11088	229	12	https://doi.org/10.1016/j.jcp.2016.10.070	https://doi.org/10.1016/j.jcp.2016.10.070	ADP
app01-11088	229	13	https://doi.org/10.1016/j.pmatsci.2018.01.005	https://doi.org/10.1016/j.pmatsci.2018.01.005	PROPN
app01-11088	229	14	https://doi.org/10.1016/j.cma.2023.115901	https://doi.org/10.1016/j.cma.2023.115901	PROPN
app01-11088	229	15	https://doi.org/10.1115/1.4029768	https://doi.org/10.1115/1.4029768	PROPN
app01-11088	229	16	https://doi.org/10.1109/tit.1962.1057692	https://doi.org/10.1109/tit.1962.1057692	PROPN
app01-11088	229	17	https://doi.org/10.1016/0031-3203(93)90098-h	https://doi.org/10.1016/0031-3203(93)90098-h	PROPN
app01-11088	229	18	https://doi.org/10.15579/gcsr.vol1.ch1	https://doi.org/10.15579/gcsr.vol1.ch1	VERB
app01-11088	229	19	https://doi.org/10.1007/978-3-319-10590-1_53	https://doi.org/10.1007/978-3-319-10590-1_53	PROPN
app01-11088	229	20	https://doi.org/10.48550/arxiv.1412.6980	https://doi.org/10.48550/arxiv.1412.6980	PROPN
app01-11088	229	21	https://doi.org/10.7717/peerj.453	https://doi.org/10.7717/peerj.453	NOUN
app01-11088	229	22	acta	acta	PROPN
app01-11088	229	23	polytechnica	polytechnica	PROPN
app01-11088	229	24	ctu	ctu	NOUN
app01-11088	229	25	proceedings	proceeding	NOUN
app01-11088	229	26	54:17–22	54:17–22	NUM
app01-11088	229	27	,	,	PUNCT
app01-11088	229	28	2025	2025	NUM
app01-11088	229	29	1	1	NUM
app01-11088	229	30	introduction	introduction	NOUN
app01-11088	229	31	2	2	NUM
app01-11088	229	32	methodology	methodology	NOUN
app01-11088	229	33	2.1	2.1	NUM
app01-11088	229	34	basic	basic	ADJ
app01-11088	229	35	theory	theory	NOUN
app01-11088	229	36	of	of	ADP
app01-11088	229	37	image	image	NOUN
app01-11088	229	38	moments	moment	NOUN
app01-11088	229	39	2.2	2.2	NUM
app01-11088	229	40	neural	neural	ADJ
app01-11088	229	41	network	network	NOUN
app01-11088	229	42	for	for	ADP
app01-11088	229	43	aggregate	aggregate	ADJ
app01-11088	229	44	generation	generation	NOUN
app01-11088	229	45	2.3	2.3	NUM
app01-11088	229	46	aggregate	aggregate	ADJ
app01-11088	229	47	dataset	dataset	VERB
app01-11088	229	48	2.4	2.4	NUM
app01-11088	229	49	training	training	NOUN
app01-11088	229	50	3	3	NUM
app01-11088	229	51	results	result	VERB
app01-11088	229	52	4	4	NUM
app01-11088	229	53	conclusion	conclusion	NOUN
app01-11088	229	54	acknowledgements	acknowledgement	NOUN
app01-11088	229	55	references	reference	NOUN