id	sid	tid	token	lemma	pos
fbem-203	1	1	frontiers	frontier	NOUN
fbem-203	1	2	in	in	ADP
fbem-203	1	3	business	business	NOUN
fbem-203	1	4	,	,	PUNCT
fbem-203	1	5	economics	economic	NOUN
fbem-203	1	6	and	and	CCONJ
fbem-203	1	7	management	management	NOUN
fbem-203	1	8	issn	issn	PROPN
fbem-203	1	9	:	:	PUNCT
fbem-203	1	10	2766	2766	NUM
fbem-203	1	11	-	-	PUNCT
fbem-203	1	12	824x	824x	NUM
fbem-203	1	13	|	|	ADJ
fbem-203	1	14	vol	vol	NOUN
fbem-203	1	15	.	.	PROPN
fbem-203	2	1	2	2	NUM
fbem-203	2	2	,	,	PUNCT
fbem-203	2	3	no	no	INTJ
fbem-203	2	4	.	.	NOUN
fbem-203	2	5	3	3	NUM
fbem-203	2	6	,	,	PUNCT
fbem-203	2	7	2021	2021	NUM
fbem-203	2	8	78	78	NUM
fbem-203	2	9	research	research	NOUN
fbem-203	2	10	on	on	ADP
fbem-203	2	11	target	target	NOUN
fbem-203	2	12	tracking	tracking	NOUN
fbem-203	2	13	algorithm	algorithm	NOUN
fbem-203	2	14	of	of	ADP
fbem-203	2	15	twin	twin	ADJ
fbem-203	2	16	networks	network	NOUN
fbem-203	2	17	integrating	integrate	VERB
fbem-203	2	18	attention	attention	NOUN
fbem-203	2	19	mechanism	mechanism	NOUN
fbem-203	2	20	relizha	relizha	NOUN
fbem-203	2	21	yeerlanbieke1	yeerlanbieke1	PROPN
fbem-203	2	22	,	,	PUNCT
fbem-203	2	23	huazhang	huazhang	PROPN
fbem-203	2	24	wang1	wang1	PROPN
fbem-203	3	1	1southwest	1southwest	NUM
fbem-203	3	2	minzu	minzu	NOUN
fbem-203	3	3	university	university	NOUN
fbem-203	3	4	college	college	NOUN
fbem-203	3	5	of	of	ADP
fbem-203	3	6	electrical	electrical	ADJ
fbem-203	3	7	engineering	engineering	NOUN
fbem-203	3	8	,	,	PUNCT
fbem-203	3	9	chengdu	chengdu	PROPN
fbem-203	3	10	,	,	PUNCT
fbem-203	3	11	sichuan	sichuan	PROPN
fbem-203	3	12	,	,	PUNCT
fbem-203	3	13	610041	610041	NUM
fbem-203	3	14	,	,	PUNCT
fbem-203	3	15	china	china	PROPN
fbem-203	3	16	abstract	abstract	NOUN
fbem-203	3	17	:	:	PUNCT
fbem-203	3	18	aiming	aim	VERB
fbem-203	3	19	at	at	ADP
fbem-203	3	20	the	the	DET
fbem-203	3	21	current	current	ADJ
fbem-203	3	22	stage	stage	NOUN
fbem-203	3	23	of	of	ADP
fbem-203	3	24	the	the	DET
fbem-203	3	25	twin	twin	ADJ
fbem-203	3	26	network	network	NOUN
fbem-203	3	27	target	target	NOUN
fbem-203	3	28	tracking	tracking	NOUN
fbem-203	3	29	algorithm	algorithm	NOUN
fbem-203	3	30	,	,	PUNCT
fbem-203	3	31	the	the	DET
fbem-203	3	32	tracking	tracking	NOUN
fbem-203	3	33	target	target	NOUN
fbem-203	3	34	is	be	AUX
fbem-203	3	35	occluded	occlude	VERB
fbem-203	3	36	,	,	PUNCT
fbem-203	3	37	the	the	DET
fbem-203	3	38	tracking	tracking	NOUN
fbem-203	3	39	is	be	AUX
fbem-203	3	40	affected	affect	VERB
fbem-203	3	41	by	by	ADP
fbem-203	3	42	illumination	illumination	NOUN
fbem-203	3	43	,	,	PUNCT
fbem-203	3	44	and	and	CCONJ
fbem-203	3	45	the	the	DET
fbem-203	3	46	target	target	NOUN
fbem-203	3	47	's	's	PART
fbem-203	3	48	scale	scale	NOUN
fbem-203	3	49	change	change	NOUN
fbem-203	3	50	from	from	ADP
fbem-203	3	51	far	far	ADV
fbem-203	3	52	to	to	ADP
fbem-203	3	53	near	near	ADJ
fbem-203	3	54	or	or	CCONJ
fbem-203	3	55	from	from	ADP
fbem-203	3	56	near	near	ADJ
fbem-203	3	57	to	to	AUX
fbem-203	3	58	far	far	ADV
fbem-203	3	59	causes	cause	VERB
fbem-203	3	60	tracking	tracking	NOUN
fbem-203	3	61	failure	failure	NOUN
fbem-203	3	62	.	.	PUNCT
fbem-203	4	1	this	this	DET
fbem-203	4	2	article	article	NOUN
fbem-203	4	3	will	will	AUX
fbem-203	4	4	optimize	optimize	VERB
fbem-203	4	5	and	and	CCONJ
fbem-203	4	6	improve	improve	VERB
fbem-203	4	7	from	from	ADP
fbem-203	4	8	two	two	NUM
fbem-203	4	9	directions	direction	NOUN
fbem-203	4	10	.	.	PUNCT
fbem-203	5	1	the	the	DET
fbem-203	5	2	twin	twin	ADJ
fbem-203	5	3	neural	neural	ADJ
fbem-203	5	4	network	network	NOUN
fbem-203	5	5	first	first	ADV
fbem-203	5	6	uses	use	VERB
fbem-203	5	7	an	an	DET
fbem-203	5	8	adaptive	adaptive	ADJ
fbem-203	5	9	detailed	detailed	ADJ
fbem-203	5	10	feature	feature	NOUN
fbem-203	5	11	extraction	extraction	NOUN
fbem-203	5	12	,	,	PUNCT
fbem-203	5	13	adds	add	VERB
fbem-203	5	14	a	a	DET
fbem-203	5	15	residual	residual	ADJ
fbem-203	5	16	network	network	NOUN
fbem-203	5	17	to	to	ADP
fbem-203	5	18	the	the	DET
fbem-203	5	19	twin	twin	ADJ
fbem-203	5	20	network	network	NOUN
fbem-203	5	21	,	,	PUNCT
fbem-203	5	22	and	and	CCONJ
fbem-203	5	23	embeds	embed	VERB
fbem-203	5	24	a	a	DET
fbem-203	5	25	detailed	detailed	ADJ
fbem-203	5	26	feature	feature	NOUN
fbem-203	5	27	retention	retention	NOUN
fbem-203	5	28	module	module	NOUN
fbem-203	5	29	in	in	ADP
fbem-203	5	30	each	each	DET
fbem-203	5	31	layer	layer	NOUN
fbem-203	5	32	,	,	PUNCT
fbem-203	5	33	amplifies	amplify	VERB
fbem-203	5	34	the	the	DET
fbem-203	5	35	changes	change	NOUN
fbem-203	5	36	in	in	ADP
fbem-203	5	37	the	the	DET
fbem-203	5	38	target	target	NOUN
fbem-203	5	39	feature	feature	NOUN
fbem-203	5	40	,	,	PUNCT
fbem-203	5	41	and	and	CCONJ
fbem-203	5	42	retains	retain	VERB
fbem-203	5	43	the	the	DET
fbem-203	5	44	important	important	ADJ
fbem-203	5	45	structure	structure	NOUN
fbem-203	5	46	of	of	ADP
fbem-203	5	47	the	the	DET
fbem-203	5	48	original	original	ADJ
fbem-203	5	49	target	target	NOUN
fbem-203	5	50	feature	feature	NOUN
fbem-203	5	51	details	detail	NOUN
fbem-203	5	52	:	:	PUNCT
fbem-203	5	53	secondly	secondly	ADV
fbem-203	5	54	,	,	PUNCT
fbem-203	5	55	the	the	DET
fbem-203	5	56	introduction	introduction	NOUN
fbem-203	5	57	of	of	ADP
fbem-203	5	58	a	a	DET
fbem-203	5	59	spatial	spatial	ADJ
fbem-203	5	60	attention	attention	NOUN
fbem-203	5	61	mechanism	mechanism	NOUN
fbem-203	5	62	allows	allow	VERB
fbem-203	5	63	the	the	DET
fbem-203	5	64	main	main	ADJ
fbem-203	5	65	branch	branch	NOUN
fbem-203	5	66	to	to	PART
fbem-203	5	67	pay	pay	VERB
fbem-203	5	68	more	more	ADJ
fbem-203	5	69	attention	attention	NOUN
fbem-203	5	70	to	to	ADP
fbem-203	5	71	the	the	DET
fbem-203	5	72	area	area	NOUN
fbem-203	5	73	to	to	PART
fbem-203	5	74	be	be	AUX
fbem-203	5	75	matched	match	VERB
fbem-203	5	76	,	,	PUNCT
fbem-203	5	77	improves	improve	VERB
fbem-203	5	78	the	the	DET
fbem-203	5	79	ability	ability	NOUN
fbem-203	5	80	to	to	PART
fbem-203	5	81	distinguish	distinguish	VERB
fbem-203	5	82	features	feature	NOUN
fbem-203	5	83	,	,	PUNCT
fbem-203	5	84	and	and	CCONJ
fbem-203	5	85	makes	make	VERB
fbem-203	5	86	the	the	DET
fbem-203	5	87	tracking	tracking	NOUN
fbem-203	5	88	effect	effect	NOUN
fbem-203	5	89	better	well	ADV
fbem-203	5	90	.	.	PUNCT
fbem-203	6	1	in	in	ADP
fbem-203	6	2	order	order	NOUN
fbem-203	6	3	to	to	PART
fbem-203	6	4	verify	verify	VERB
fbem-203	6	5	the	the	DET
fbem-203	6	6	effectiveness	effectiveness	NOUN
fbem-203	6	7	of	of	ADP
fbem-203	6	8	this	this	DET
fbem-203	6	9	experiment	experiment	NOUN
fbem-203	6	10	,	,	PUNCT
fbem-203	6	11	this	this	DET
fbem-203	6	12	experiment	experiment	NOUN
fbem-203	6	13	was	be	AUX
fbem-203	6	14	tested	test	VERB
fbem-203	6	15	on	on	ADP
fbem-203	6	16	the	the	DET
fbem-203	6	17	data	datum	NOUN
fbem-203	6	18	set	set	VERB
fbem-203	6	19	otb2015	otb2015	NOUN
fbem-203	6	20	.	.	PUNCT
fbem-203	7	1	the	the	DET
fbem-203	7	2	experiment	experiment	NOUN
fbem-203	7	3	proved	prove	VERB
fbem-203	7	4	that	that	SCONJ
fbem-203	7	5	the	the	DET
fbem-203	7	6	proposed	propose	VERB
fbem-203	7	7	algorithm	algorithm	NOUN
fbem-203	7	8	performs	perform	VERB
fbem-203	7	9	better	well	ADV
fbem-203	7	10	in	in	ADP
fbem-203	7	11	accuracy	accuracy	NOUN
fbem-203	7	12	and	and	CCONJ
fbem-203	7	13	success	success	NOUN
fbem-203	7	14	rate	rate	NOUN
fbem-203	7	15	.	.	PUNCT
fbem-203	8	1	keywords	keyword	NOUN
fbem-203	8	2	:	:	PUNCT
fbem-203	8	3	deep	deep	ADJ
fbem-203	8	4	learning	learning	NOUN
fbem-203	8	5	,	,	PUNCT
fbem-203	8	6	twin	twin	ADJ
fbem-203	8	7	neural	neural	ADJ
fbem-203	8	8	network	network	NOUN
fbem-203	8	9	,	,	PUNCT
fbem-203	8	10	target	target	NOUN
fbem-203	8	11	tracking	tracking	NOUN
fbem-203	8	12	,	,	PUNCT
fbem-203	8	13	attention	attention	NOUN
fbem-203	8	14	mechanism	mechanism	NOUN
fbem-203	8	15	.	.	PUNCT
fbem-203	9	1	1	1	X
fbem-203	9	2	.	.	X
fbem-203	9	3	preface	preface	NOUN
fbem-203	9	4	target	target	NOUN
fbem-203	9	5	tracking	tracking	NOUN
fbem-203	9	6	technology	technology	NOUN
fbem-203	9	7	is	be	AUX
fbem-203	9	8	one	one	NUM
fbem-203	9	9	of	of	ADP
fbem-203	9	10	the	the	DET
fbem-203	9	11	important	important	ADJ
fbem-203	9	12	research	research	NOUN
fbem-203	9	13	parts	part	NOUN
fbem-203	9	14	of	of	ADP
fbem-203	9	15	artificial	artificial	ADJ
fbem-203	9	16	intelligence	intelligence	NOUN
fbem-203	9	17	research	research	NOUN
fbem-203	9	18	.	.	PUNCT
fbem-203	10	1	target	target	NOUN
fbem-203	10	2	tracking	tracking	NOUN
fbem-203	10	3	refers	refer	VERB
fbem-203	10	4	to	to	ADP
fbem-203	10	5	determining	determine	VERB
fbem-203	10	6	the	the	DET
fbem-203	10	7	coordinates	coordinate	NOUN
fbem-203	10	8	and	and	CCONJ
fbem-203	10	9	size	size	NOUN
fbem-203	10	10	of	of	ADP
fbem-203	10	11	the	the	DET
fbem-203	10	12	target	target	NOUN
fbem-203	10	13	in	in	ADP
fbem-203	10	14	the	the	DET
fbem-203	10	15	initial	initial	ADJ
fbem-203	10	16	frame	frame	NOUN
fbem-203	10	17	,	,	PUNCT
fbem-203	10	18	and	and	CCONJ
fbem-203	10	19	letting	let	VERB
fbem-203	10	20	the	the	DET
fbem-203	10	21	tracking	tracking	NOUN
fbem-203	10	22	frame	frame	NOUN
fbem-203	10	23	describe	describe	VERB
fbem-203	10	24	the	the	DET
fbem-203	10	25	moving	move	VERB
fbem-203	10	26	trajectory	trajectory	NOUN
fbem-203	10	27	of	of	ADP
fbem-203	10	28	the	the	DET
fbem-203	10	29	tracking	tracking	NOUN
fbem-203	10	30	target	target	NOUN
fbem-203	10	31	in	in	ADP
fbem-203	10	32	the	the	DET
fbem-203	10	33	subsequent	subsequent	ADJ
fbem-203	10	34	frames	frame	NOUN
fbem-203	10	35	.	.	PUNCT
fbem-203	11	1	now	now	ADV
fbem-203	11	2	its	its	PRON
fbem-203	11	3	technology	technology	NOUN
fbem-203	11	4	is	be	AUX
fbem-203	11	5	applied	apply	VERB
fbem-203	11	6	to	to	ADP
fbem-203	11	7	many	many	ADJ
fbem-203	11	8	fields	field	NOUN
fbem-203	11	9	such	such	ADJ
fbem-203	11	10	as	as	ADP
fbem-203	11	11	safety	safety	NOUN
fbem-203	11	12	monitoring	monitoring	NOUN
fbem-203	11	13	,	,	PUNCT
fbem-203	11	14	competitive	competitive	ADJ
fbem-203	11	15	sports	sport	NOUN
fbem-203	11	16	,	,	PUNCT
fbem-203	11	17	human	human	NOUN
fbem-203	11	18	-	-	PUNCT
fbem-203	11	19	computer	computer	NOUN
fbem-203	11	20	interaction	interaction	NOUN
fbem-203	11	21	,	,	PUNCT
fbem-203	11	22	etc	etc	X
fbem-203	11	23	.	.	X
fbem-203	12	1	[	[	X
fbem-203	12	2	1	1	NUM
fbem-203	12	3	]	]	PUNCT
fbem-203	12	4	.	.	PUNCT
fbem-203	13	1	due	due	ADP
fbem-203	13	2	to	to	ADP
fbem-203	13	3	the	the	DET
fbem-203	13	4	fast	fast	ADJ
fbem-203	13	5	speed	speed	NOUN
fbem-203	13	6	of	of	ADP
fbem-203	13	7	the	the	DET
fbem-203	13	8	twin	twin	ADJ
fbem-203	13	9	network	network	NOUN
fbem-203	13	10	target	target	NOUN
fbem-203	13	11	tracking	tracking	NOUN
fbem-203	13	12	algorithm	algorithm	NOUN
fbem-203	13	13	,	,	PUNCT
fbem-203	13	14	it	it	PRON
fbem-203	13	15	has	have	AUX
fbem-203	13	16	attracted	attract	VERB
fbem-203	13	17	the	the	DET
fbem-203	13	18	interest	interest	NOUN
fbem-203	13	19	of	of	ADP
fbem-203	13	20	many	many	ADJ
fbem-203	13	21	researchers	researcher	NOUN
fbem-203	13	22	.	.	PUNCT
fbem-203	14	1	the	the	DET
fbem-203	14	2	target	target	NOUN
fbem-203	14	3	tracking	tracking	NOUN
fbem-203	14	4	algorithm	algorithm	NOUN
fbem-203	14	5	based	base	VERB
fbem-203	14	6	on	on	ADP
fbem-203	14	7	this	this	DET
fbem-203	14	8	algorithm	algorithm	NOUN
fbem-203	14	9	has	have	AUX
fbem-203	14	10	also	also	ADV
fbem-203	14	11	developed	develop	VERB
fbem-203	14	12	rapidly	rapidly	ADV
fbem-203	14	13	.	.	PUNCT
fbem-203	15	1	the	the	DET
fbem-203	15	2	algorithm	algorithm	NOUN
fbem-203	15	3	extracts	extract	VERB
fbem-203	15	4	features	feature	NOUN
fbem-203	15	5	from	from	ADP
fbem-203	15	6	the	the	DET
fbem-203	15	7	input	input	NOUN
fbem-203	15	8	image	image	NOUN
fbem-203	15	9	and	and	CCONJ
fbem-203	15	10	the	the	DET
fbem-203	15	11	image	image	NOUN
fbem-203	15	12	to	to	PART
fbem-203	15	13	be	be	AUX
fbem-203	15	14	measured	measure	VERB
fbem-203	15	15	,	,	PUNCT
fbem-203	15	16	and	and	CCONJ
fbem-203	15	17	then	then	ADV
fbem-203	15	18	performs	perform	VERB
fbem-203	15	19	calculations	calculation	NOUN
fbem-203	15	20	and	and	CCONJ
fbem-203	15	21	evaluates	evaluate	VERB
fbem-203	15	22	the	the	DET
fbem-203	15	23	results	result	NOUN
fbem-203	15	24	of	of	ADP
fbem-203	15	25	the	the	DET
fbem-203	15	26	calculation	calculation	NOUN
fbem-203	15	27	.	.	PUNCT
fbem-203	16	1	their	their	PRON
fbem-203	16	2	similarity	similarity	NOUN
fbem-203	16	3	.	.	PUNCT
fbem-203	17	1	bertinetto	bertinetto	PROPN
fbem-203	17	2	et	et	PROPN
fbem-203	17	3	al	al	PROPN
fbem-203	17	4	.	.	PUNCT
fbem-203	18	1	[	[	X
fbem-203	18	2	2	2	X
fbem-203	18	3	]	]	PUNCT
fbem-203	18	4	proposed	propose	VERB
fbem-203	18	5	a	a	DET
fbem-203	18	6	fully	fully	ADV
fbem-203	18	7	convolutional	convolutional	ADJ
fbem-203	18	8	twin	twin	ADJ
fbem-203	18	9	network	network	NOUN
fbem-203	18	10	algorithm	algorithm	NOUN
fbem-203	18	11	.	.	PUNCT
fbem-203	19	1	the	the	DET
fbem-203	19	2	network	network	NOUN
fbem-203	19	3	has	have	VERB
fbem-203	19	4	only	only	ADV
fbem-203	19	5	two	two	NUM
fbem-203	19	6	layers	layer	NOUN
fbem-203	19	7	,	,	PUNCT
fbem-203	19	8	a	a	DET
fbem-203	19	9	convolutional	convolutional	ADJ
fbem-203	19	10	layer	layer	NOUN
fbem-203	19	11	and	and	CCONJ
fbem-203	19	12	a	a	DET
fbem-203	19	13	pooling	pool	VERB
fbem-203	19	14	layer	layer	NOUN
fbem-203	19	15	.	.	PUNCT
fbem-203	20	1	it	it	PRON
fbem-203	20	2	is	be	AUX
fbem-203	20	3	characterized	characterize	VERB
fbem-203	20	4	by	by	ADP
fbem-203	20	5	an	an	DET
fbem-203	20	6	online	online	ADJ
fbem-203	20	7	non	non	ADJ
fbem-203	20	8	-	-	ADJ
fbem-203	20	9	update	update	ADJ
fbem-203	20	10	strategy	strategy	NOUN
fbem-203	20	11	,	,	PUNCT
fbem-203	20	12	which	which	PRON
fbem-203	20	13	makes	make	VERB
fbem-203	20	14	the	the	DET
fbem-203	20	15	tracking	tracking	NOUN
fbem-203	20	16	speed	speed	NOUN
fbem-203	20	17	faster	fast	ADV
fbem-203	20	18	.	.	PUNCT
fbem-203	21	1	guo	guo	PROPN
fbem-203	21	2	et	et	PROPN
fbem-203	21	3	al	al	PROPN
fbem-203	21	4	.	.	PUNCT
fbem-203	22	1	[	[	X
fbem-203	22	2	3	3	X
fbem-203	22	3	]	]	PUNCT
fbem-203	22	4	proposed	propose	VERB
fbem-203	22	5	a	a	DET
fbem-203	22	6	dynamic	dynamic	ADJ
fbem-203	22	7	twin	twin	ADJ
fbem-203	22	8	network	network	NOUN
fbem-203	22	9	tracking	tracking	NOUN
fbem-203	22	10	algorithm	algorithm	NOUN
fbem-203	22	11	,	,	PUNCT
fbem-203	22	12	which	which	PRON
fbem-203	22	13	effectively	effectively	ADV
fbem-203	22	14	suppressed	suppress	VERB
fbem-203	22	15	the	the	DET
fbem-203	22	16	background	background	NOUN
fbem-203	22	17	interference	interference	NOUN
fbem-203	22	18	information	information	NOUN
fbem-203	22	19	and	and	CCONJ
fbem-203	22	20	the	the	DET
fbem-203	22	21	interference	interference	NOUN
fbem-203	22	22	caused	cause	VERB
fbem-203	22	23	by	by	ADP
fbem-203	22	24	deformation	deformation	NOUN
fbem-203	22	25	in	in	ADP
fbem-203	22	26	a	a	DET
fbem-203	22	27	complex	complex	ADJ
fbem-203	22	28	environment	environment	NOUN
fbem-203	22	29	,	,	PUNCT
fbem-203	22	30	and	and	CCONJ
fbem-203	22	31	effectively	effectively	ADV
fbem-203	22	32	improved	improve	VERB
fbem-203	22	33	the	the	DET
fbem-203	22	34	tracking	tracking	NOUN
fbem-203	22	35	effect	effect	NOUN
fbem-203	22	36	.	.	PUNCT
fbem-203	23	1	li	li	PROPN
fbem-203	23	2	et	et	PROPN
fbem-203	23	3	al	al	PROPN
fbem-203	23	4	.	.	PUNCT
fbem-203	24	1	[	[	X
fbem-203	24	2	4	4	X
fbem-203	24	3	]	]	PUNCT
fbem-203	24	4	proposed	propose	VERB
fbem-203	24	5	an	an	DET
fbem-203	24	6	efficient	efficient	ADJ
fbem-203	24	7	twin	twin	ADJ
fbem-203	24	8	area	area	NOUN
fbem-203	24	9	recommendation	recommendation	NOUN
fbem-203	24	10	network	network	NOUN
fbem-203	24	11	tracking	tracking	NOUN
fbem-203	24	12	algorithm	algorithm	NOUN
fbem-203	24	13	,	,	PUNCT
fbem-203	24	14	which	which	PRON
fbem-203	24	15	merged	merge	VERB
fbem-203	24	16	the	the	DET
fbem-203	24	17	area	area	NOUN
fbem-203	24	18	recommendation	recommendation	NOUN
fbem-203	24	19	network	network	NOUN
fbem-203	24	20	into	into	ADP
fbem-203	24	21	the	the	DET
fbem-203	24	22	twin	twin	ADJ
fbem-203	24	23	neural	neural	ADJ
fbem-203	24	24	network	network	NOUN
fbem-203	24	25	structure	structure	NOUN
fbem-203	24	26	.	.	PUNCT
fbem-203	25	1	rpn	rpn	PROPN
fbem-203	25	2	is	be	AUX
fbem-203	25	3	composed	compose	VERB
fbem-203	25	4	of	of	ADP
fbem-203	25	5	classification	classification	NOUN
fbem-203	25	6	branches	branch	NOUN
fbem-203	25	7	and	and	CCONJ
fbem-203	25	8	regression	regression	NOUN
fbem-203	25	9	branches	branch	NOUN
fbem-203	25	10	.	.	PUNCT
fbem-203	26	1	the	the	DET
fbem-203	26	2	classification	classification	NOUN
fbem-203	26	3	branch	branch	NOUN
fbem-203	26	4	is	be	AUX
fbem-203	26	5	used	use	VERB
fbem-203	26	6	to	to	PART
fbem-203	26	7	identify	identify	VERB
fbem-203	26	8	the	the	DET
fbem-203	26	9	target	target	NOUN
fbem-203	26	10	and	and	CCONJ
fbem-203	26	11	background	background	NOUN
fbem-203	26	12	;	;	PUNCT
fbem-203	26	13	the	the	DET
fbem-203	26	14	regression	regression	NOUN
fbem-203	26	15	branch	branch	NOUN
fbem-203	26	16	is	be	AUX
fbem-203	26	17	used	use	VERB
fbem-203	26	18	to	to	PART
fbem-203	26	19	adjust	adjust	VERB
fbem-203	26	20	the	the	DET
fbem-203	26	21	candidate	candidate	NOUN
fbem-203	26	22	area	area	NOUN
fbem-203	26	23	.	.	PUNCT
fbem-203	27	1	zhu	zhu	PROPN
fbem-203	27	2	et	et	PROPN
fbem-203	27	3	al	al	PROPN
fbem-203	27	4	.	.	PUNCT
fbem-203	28	1	[	[	X
fbem-203	28	2	5	5	NUM
fbem-203	28	3	]	]	PUNCT
fbem-203	28	4	proposed	propose	VERB
fbem-203	28	5	a	a	DET
fbem-203	28	6	twin	twin	ADJ
fbem-203	28	7	network	network	NOUN
fbem-203	28	8	tracking	tracking	NOUN
fbem-203	28	9	algorithm	algorithm	NOUN
fbem-203	28	10	based	base	VERB
fbem-203	28	11	on	on	ADP
fbem-203	28	12	jammer	jammer	NOUN
fbem-203	28	13	perception	perception	NOUN
fbem-203	28	14	.	.	PUNCT
fbem-203	29	1	compared	compare	VERB
fbem-203	29	2	with	with	ADP
fbem-203	29	3	siamrpn	siamrpn	NOUN
fbem-203	29	4	,	,	PUNCT
fbem-203	29	5	this	this	DET
fbem-203	29	6	algorithm	algorithm	NOUN
fbem-203	29	7	can	can	AUX
fbem-203	29	8	not	not	PART
fbem-203	29	9	only	only	ADV
fbem-203	29	10	get	get	VERB
fbem-203	29	11	a	a	DET
fbem-203	29	12	better	well	ADJ
fbem-203	29	13	tracking	tracking	NOUN
fbem-203	29	14	frame	frame	NOUN
fbem-203	29	15	in	in	ADP
fbem-203	29	16	the	the	DET
fbem-203	29	17	tracking	tracking	NOUN
fbem-203	29	18	process	process	NOUN
fbem-203	29	19	,	,	PUNCT
fbem-203	29	20	but	but	CCONJ
fbem-203	29	21	also	also	ADV
fbem-203	29	22	improve	improve	VERB
fbem-203	29	23	the	the	DET
fbem-203	29	24	generalization	generalization	NOUN
fbem-203	29	25	ability	ability	NOUN
fbem-203	29	26	of	of	ADP
fbem-203	29	27	the	the	DET
fbem-203	29	28	tracker	tracker	NOUN
fbem-203	30	1	[	[	X
fbem-203	30	2	6	6	NUM
fbem-203	30	3	]	]	PUNCT
fbem-203	30	4	the	the	DET
fbem-203	30	5	role	role	NOUN
fbem-203	30	6	of	of	ADP
fbem-203	30	7	.	.	PUNCT
fbem-203	31	1	wang	wang	PROPN
fbem-203	31	2	et	et	PROPN
fbem-203	31	3	al	al	PROPN
fbem-203	31	4	.	.	PUNCT
fbem-203	32	1	[	[	X
fbem-203	32	2	7	7	X
fbem-203	32	3	]	]	PUNCT
fbem-203	32	4	proposed	propose	VERB
fbem-203	32	5	a	a	DET
fbem-203	32	6	fast	fast	ADJ
fbem-203	32	7	online	online	NOUN
fbem-203	32	8	target	target	NOUN
fbem-203	32	9	tracking	tracking	NOUN
fbem-203	32	10	and	and	CCONJ
fbem-203	32	11	segmentation	segmentation	NOUN
fbem-203	32	12	algorithm	algorithm	NOUN
fbem-203	33	1	[	[	X
fbem-203	33	2	8	8	NUM
fbem-203	33	3	]	]	PUNCT
fbem-203	33	4	,	,	PUNCT
fbem-203	33	5	mainly	mainly	ADV
fbem-203	33	6	replacing	replace	VERB
fbem-203	33	7	alexnet	alexnet	NOUN
fbem-203	33	8	with	with	ADP
fbem-203	33	9	resnet	resnet	NOUN
fbem-203	33	10	[	[	X
fbem-203	33	11	9	9	NUM
fbem-203	33	12	]	]	PUNCT
fbem-203	33	13	and	and	CCONJ
fbem-203	33	14	adding	add	VERB
fbem-203	33	15	mask	mask	NOUN
fbem-203	33	16	to	to	ADP
fbem-203	33	17	rpn	rpn	PROPN
fbem-203	33	18	in	in	ADP
fbem-203	33	19	parallel	parallel	NOUN
fbem-203	33	20	to	to	PART
fbem-203	33	21	further	far	ADV
fbem-203	33	22	improve	improve	VERB
fbem-203	33	23	the	the	DET
fbem-203	33	24	accuracy	accuracy	NOUN
fbem-203	33	25	of	of	ADP
fbem-203	33	26	tracking	tracking	NOUN
fbem-203	33	27	.	.	PUNCT
fbem-203	34	1	it	it	PRON
fbem-203	34	2	can	can	AUX
fbem-203	34	3	be	be	AUX
fbem-203	34	4	concluded	conclude	VERB
fbem-203	34	5	that	that	SCONJ
fbem-203	34	6	the	the	DET
fbem-203	34	7	existing	exist	VERB
fbem-203	34	8	tracking	tracking	NOUN
fbem-203	34	9	algorithm	algorithm	NOUN
fbem-203	34	10	based	base	VERB
fbem-203	34	11	on	on	ADP
fbem-203	34	12	the	the	DET
fbem-203	34	13	twin	twin	ADJ
fbem-203	34	14	network	network	NOUN
fbem-203	34	15	performs	perform	VERB
fbem-203	34	16	very	very	ADV
fbem-203	34	17	well	well	ADV
fbem-203	34	18	in	in	ADP
fbem-203	34	19	real	real	ADJ
fbem-203	34	20	-	-	PUNCT
fbem-203	34	21	time	time	NOUN
fbem-203	34	22	,	,	PUNCT
fbem-203	34	23	but	but	CCONJ
fbem-203	34	24	in	in	ADP
fbem-203	34	25	the	the	DET
fbem-203	34	26	tracking	tracking	NOUN
fbem-203	34	27	process	process	NOUN
fbem-203	34	28	,	,	PUNCT
fbem-203	34	29	the	the	DET
fbem-203	34	30	extraction	extraction	NOUN
fbem-203	34	31	of	of	ADP
fbem-203	34	32	the	the	DET
fbem-203	34	33	target	target	NOUN
fbem-203	34	34	feature	feature	NOUN
fbem-203	34	35	is	be	AUX
fbem-203	34	36	often	often	ADV
fbem-203	34	37	based	base	VERB
fbem-203	34	38	on	on	ADP
fbem-203	34	39	the	the	DET
fbem-203	34	40	preprocessed	preprocesse	VERB
fbem-203	34	41	first	first	ADJ
fbem-203	34	42	frame	frame	NOUN
fbem-203	34	43	of	of	ADP
fbem-203	34	44	picture	picture	NOUN
fbem-203	34	45	,	,	PUNCT
fbem-203	34	46	and	and	CCONJ
fbem-203	34	47	no	no	DET
fbem-203	34	48	further	further	ADJ
fbem-203	34	49	updates	update	NOUN
fbem-203	34	50	will	will	AUX
fbem-203	34	51	be	be	AUX
fbem-203	34	52	performed	perform	VERB
fbem-203	34	53	afterwards	afterwards	ADV
fbem-203	34	54	.	.	PUNCT
fbem-203	35	1	therefore	therefore	ADV
fbem-203	35	2	,	,	PUNCT
fbem-203	35	3	when	when	SCONJ
fbem-203	35	4	the	the	DET
fbem-203	35	5	target	target	NOUN
fbem-203	35	6	's	's	PART
fbem-203	35	7	shape	shape	NOUN
fbem-203	35	8	changes	change	NOUN
fbem-203	35	9	from	from	ADP
fbem-203	35	10	far	far	ADV
fbem-203	35	11	to	to	ADP
fbem-203	35	12	near	near	ADJ
fbem-203	35	13	or	or	CCONJ
fbem-203	35	14	from	from	ADP
fbem-203	35	15	near	near	ADJ
fbem-203	35	16	to	to	ADP
fbem-203	35	17	far	far	ADV
fbem-203	35	18	,	,	PUNCT
fbem-203	35	19	it	it	PRON
fbem-203	35	20	can	can	AUX
fbem-203	35	21	not	not	PART
fbem-203	35	22	better	well	ADV
fbem-203	35	23	adapt	adapt	VERB
fbem-203	35	24	to	to	ADP
fbem-203	35	25	the	the	DET
fbem-203	35	26	changes	change	NOUN
fbem-203	35	27	of	of	ADP
fbem-203	35	28	the	the	DET
fbem-203	35	29	target	target	NOUN
fbem-203	35	30	;	;	PUNCT
fbem-203	35	31	when	when	SCONJ
fbem-203	35	32	the	the	DET
fbem-203	35	33	background	background	NOUN
fbem-203	35	34	and	and	CCONJ
fbem-203	35	35	foreground	foreground	VERB
fbem-203	35	36	in	in	ADP
fbem-203	35	37	the	the	DET
fbem-203	35	38	template	template	NOUN
fbem-203	35	39	image	image	NOUN
fbem-203	35	40	are	be	AUX
fbem-203	35	41	similar	similar	ADJ
fbem-203	35	42	in	in	ADP
fbem-203	35	43	appearance	appearance	NOUN
fbem-203	35	44	,	,	PUNCT
fbem-203	35	45	the	the	DET
fbem-203	35	46	background	background	NOUN
fbem-203	35	47	image	image	NOUN
fbem-203	35	48	may	may	AUX
fbem-203	35	49	get	get	VERB
fbem-203	35	50	a	a	DET
fbem-203	35	51	larger	large	ADJ
fbem-203	35	52	similarity	similarity	NOUN
fbem-203	35	53	score	score	NOUN
fbem-203	35	54	.	.	PUNCT
fbem-203	36	1	lead	lead	VERB
fbem-203	36	2	to	to	ADP
fbem-203	36	3	tracking	track	VERB
fbem-203	36	4	drift	drift	NOUN
fbem-203	36	5	and	and	CCONJ
fbem-203	36	6	other	other	ADJ
fbem-203	36	7	problems	problem	NOUN
fbem-203	36	8	.	.	PUNCT
fbem-203	37	1	aiming	aim	VERB
fbem-203	37	2	at	at	ADP
fbem-203	37	3	the	the	DET
fbem-203	37	4	above	above	ADV
fbem-203	37	5	-	-	PUNCT
fbem-203	37	6	mentioned	mention	VERB
fbem-203	37	7	problems	problem	NOUN
fbem-203	37	8	based	base	VERB
fbem-203	37	9	on	on	ADP
fbem-203	37	10	the	the	DET
fbem-203	37	11	twin	twin	ADJ
fbem-203	37	12	neural	neural	ADJ
fbem-203	37	13	network	network	NOUN
fbem-203	37	14	tracking	tracking	NOUN
fbem-203	37	15	algorithm	algorithm	NOUN
fbem-203	37	16	,	,	PUNCT
fbem-203	37	17	this	this	DET
fbem-203	37	18	article	article	NOUN
fbem-203	37	19	made	make	VERB
fbem-203	37	20	the	the	DET
fbem-203	37	21	following	follow	VERB
fbem-203	37	22	improvements	improvement	NOUN
fbem-203	37	23	on	on	ADP
fbem-203	37	24	the	the	DET
fbem-203	37	25	twin	twin	ADJ
fbem-203	37	26	network	network	NOUN
fbem-203	37	27	algorithm	algorithm	NOUN
fbem-203	37	28	:	:	PUNCT
fbem-203	37	29	1	1	X
fbem-203	37	30	)	)	PUNCT
fbem-203	37	31	using	use	VERB
fbem-203	37	32	an	an	DET
fbem-203	37	33	adaptive	adaptive	ADJ
fbem-203	37	34	detailed	detailed	ADJ
fbem-203	37	35	feature	feature	NOUN
fbem-203	37	36	extraction	extraction	NOUN
fbem-203	37	37	,	,	PUNCT
fbem-203	37	38	adding	add	VERB
fbem-203	37	39	a	a	DET
fbem-203	37	40	residual	residual	ADJ
fbem-203	37	41	network	network	NOUN
fbem-203	37	42	to	to	ADP
fbem-203	37	43	the	the	DET
fbem-203	37	44	twin	twin	ADJ
fbem-203	37	45	network	network	NOUN
fbem-203	37	46	,	,	PUNCT
fbem-203	37	47	and	and	CCONJ
fbem-203	37	48	the	the	DET
fbem-203	37	49	detail	detail	NOUN
fbem-203	37	50	feature	feature	NOUN
fbem-203	37	51	retention	retention	NOUN
fbem-203	37	52	module	module	NOUN
fbem-203	37	53	is	be	AUX
fbem-203	37	54	embedded	embed	VERB
fbem-203	37	55	in	in	ADP
fbem-203	37	56	each	each	DET
fbem-203	37	57	layer	layer	NOUN
fbem-203	37	58	to	to	PART
fbem-203	37	59	amplify	amplify	VERB
fbem-203	37	60	the	the	DET
fbem-203	37	61	changes	change	NOUN
fbem-203	37	62	of	of	ADP
fbem-203	37	63	target	target	NOUN
fbem-203	37	64	features	feature	NOUN
fbem-203	37	65	and	and	CCONJ
fbem-203	37	66	retain	retain	VERB
fbem-203	37	67	the	the	DET
fbem-203	37	68	important	important	ADJ
fbem-203	37	69	structural	structural	ADJ
fbem-203	37	70	details	detail	NOUN
fbem-203	37	71	of	of	ADP
fbem-203	37	72	the	the	DET
fbem-203	37	73	original	original	ADJ
fbem-203	37	74	target	target	NOUN
fbem-203	37	75	features	feature	NOUN
fbem-203	37	76	.	.	PUNCT
fbem-203	38	1	it	it	PRON
fbem-203	38	2	plays	play	VERB
fbem-203	38	3	a	a	DET
fbem-203	38	4	certain	certain	ADJ
fbem-203	38	5	role	role	NOUN
fbem-203	38	6	in	in	ADP
fbem-203	38	7	the	the	DET
fbem-203	38	8	search	search	NOUN
fbem-203	38	9	and	and	CCONJ
fbem-203	38	10	discrimination	discrimination	NOUN
fbem-203	38	11	of	of	ADP
fbem-203	38	12	details	detail	NOUN
fbem-203	38	13	in	in	ADP
fbem-203	38	14	target	target	NOUN
fbem-203	38	15	tracking	tracking	NOUN
fbem-203	38	16	,	,	PUNCT
fbem-203	38	17	which	which	PRON
fbem-203	38	18	solves	solve	VERB
fbem-203	38	19	the	the	DET
fbem-203	38	20	problem	problem	NOUN
fbem-203	38	21	of	of	ADP
fbem-203	38	22	the	the	DET
fbem-203	38	23	original	original	ADJ
fbem-203	38	24	tracking	tracking	NOUN
fbem-203	38	25	network	network	NOUN
fbem-203	38	26	in	in	ADP
fbem-203	38	27	the	the	DET
fbem-203	38	28	parameter	parameter	NOUN
fbem-203	38	29	scale	scale	NOUN
fbem-203	38	30	.	.	PUNCT
fbem-203	39	1	the	the	DET
fbem-203	39	2	problem	problem	NOUN
fbem-203	39	3	of	of	ADP
fbem-203	39	4	the	the	DET
fbem-203	39	5	loss	loss	NOUN
fbem-203	39	6	of	of	ADP
fbem-203	39	7	detailed	detailed	ADJ
fbem-203	39	8	features	feature	NOUN
fbem-203	39	9	of	of	ADP
fbem-203	39	10	the	the	DET
fbem-203	39	11	target	target	NOUN
fbem-203	39	12	when	when	SCONJ
fbem-203	39	13	it	it	PRON
fbem-203	39	14	is	be	AUX
fbem-203	39	15	lowered	lower	VERB
fbem-203	39	16	.	.	PUNCT
fbem-203	40	1	2	2	X
fbem-203	40	2	)	)	PUNCT
fbem-203	40	3	the	the	DET
fbem-203	40	4	spatial	spatial	ADJ
fbem-203	40	5	attention	attention	NOUN
fbem-203	40	6	mechanism	mechanism	NOUN
fbem-203	40	7	is	be	AUX
fbem-203	40	8	used	use	VERB
fbem-203	40	9	to	to	PART
fbem-203	40	10	improve	improve	VERB
fbem-203	40	11	the	the	DET
fbem-203	40	12	network	network	NOUN
fbem-203	40	13	model	model	NOUN
fbem-203	40	14	's	's	PART
fbem-203	40	15	attention	attention	NOUN
fbem-203	40	16	to	to	ADP
fbem-203	40	17	effective	effective	ADJ
fbem-203	40	18	features	feature	NOUN
fbem-203	40	19	during	during	ADP
fbem-203	40	20	feature	feature	NOUN
fbem-203	40	21	extraction	extraction	NOUN
fbem-203	40	22	.	.	PUNCT
fbem-203	41	1	2	2	X
fbem-203	41	2	.	.	X
fbem-203	41	3	related	relate	VERB
fbem-203	41	4	work	work	NOUN
fbem-203	41	5	2.1	2.1	NUM
fbem-203	41	6	.	.	PUNCT
fbem-203	42	1	target	target	NOUN
fbem-203	42	2	tracking	tracking	NOUN
fbem-203	42	3	algorithm	algorithm	NOUN
fbem-203	42	4	based	base	VERB
fbem-203	42	5	on	on	ADP
fbem-203	42	6	twin	twin	ADJ
fbem-203	42	7	neural	neural	ADJ
fbem-203	42	8	network	network	NOUN
fbem-203	42	9	the	the	DET
fbem-203	42	10	siamfc	siamfc	NOUN
fbem-203	42	11	network	network	NOUN
fbem-203	42	12	is	be	AUX
fbem-203	42	13	mainly	mainly	ADV
fbem-203	42	14	used	use	VERB
fbem-203	42	15	to	to	PART
fbem-203	42	16	calculate	calculate	VERB
fbem-203	42	17	the	the	DET
fbem-203	42	18	similarity	similarity	NOUN
fbem-203	42	19	of	of	ADP
fbem-203	42	20	the	the	DET
fbem-203	42	21	features	feature	NOUN
fbem-203	42	22	extracted	extract	VERB
fbem-203	42	23	by	by	ADP
fbem-203	42	24	the	the	DET
fbem-203	42	25	two	two	NUM
fbem-203	42	26	branches	branch	NOUN
fbem-203	42	27	in	in	ADP
fbem-203	42	28	the	the	DET
fbem-203	42	29	twin	twin	ADJ
fbem-203	42	30	network	network	NOUN
fbem-203	42	31	,	,	PUNCT
fbem-203	42	32	and	and	CCONJ
fbem-203	42	33	estimate	estimate	VERB
fbem-203	42	34	the	the	DET
fbem-203	42	35	position	position	NOUN
fbem-203	42	36	of	of	ADP
fbem-203	42	37	the	the	DET
fbem-203	42	38	target	target	NOUN
fbem-203	42	39	through	through	ADP
fbem-203	42	40	the	the	DET
fbem-203	42	41	highest	high	ADJ
fbem-203	42	42	point	point	NOUN
fbem-203	42	43	of	of	ADP
fbem-203	42	44	the	the	DET
fbem-203	42	45	similarity	similarity	NOUN
fbem-203	42	46	response	response	NOUN
fbem-203	42	47	.	.	PUNCT
fbem-203	43	1	the	the	DET
fbem-203	43	2	target	target	NOUN
fbem-203	43	3	tracking	tracking	NOUN
fbem-203	43	4	process	process	NOUN
fbem-203	43	5	can	can	AUX
fbem-203	43	6	be	be	AUX
fbem-203	43	7	regarded	regard	VERB
fbem-203	43	8	as	as	ADP
fbem-203	43	9	a	a	DET
fbem-203	43	10	method	method	NOUN
fbem-203	43	11	to	to	PART
fbem-203	43	12	solve	solve	VERB
fbem-203	43	13	similar	similar	ADJ
fbem-203	43	14	learning	learning	NOUN
fbem-203	43	15	problems	problem	NOUN
fbem-203	43	16	[	[	X
fbem-203	43	17	10	10	NUM
fbem-203	43	18	]	]	PUNCT
fbem-203	43	19	.	.	PUNCT
fbem-203	44	1	if	if	SCONJ
fbem-203	44	2	the	the	DET
fbem-203	44	3	target	target	NOUN
fbem-203	44	4	image	image	NOUN
fbem-203	44	5	z	z	NOUN
fbem-203	44	6	and	and	CCONJ
fbem-203	44	7	the	the	DET
fbem-203	44	8	candidate	candidate	NOUN
fbem-203	44	9	image	image	NOUN
fbem-203	44	10	x	x	VERB
fbem-203	44	11	are	be	AUX
fbem-203	44	12	the	the	DET
fbem-203	44	13	same	same	ADJ
fbem-203	44	14	target	target	NOUN
fbem-203	44	15	,	,	PUNCT
fbem-203	44	16	then	then	ADV
fbem-203	44	17	the	the	DET
fbem-203	44	18	mapping	mapping	NOUN
fbem-203	44	19	function	function	NOUN
fbem-203	44	20	will	will	AUX
fbem-203	44	21	return	return	VERB
fbem-203	44	22	a	a	DET
fbem-203	44	23	higher	high	ADJ
fbem-203	44	24	similarity	similarity	NOUN
fbem-203	44	25	score	score	NOUN
fbem-203	44	26	,	,	PUNCT
fbem-203	44	27	on	on	ADP
fbem-203	44	28	the	the	DET
fbem-203	44	29	contrary	contrary	NOUN
fbem-203	44	30	,	,	PUNCT
fbem-203	44	31	it	it	PRON
fbem-203	44	32	will	will	AUX
fbem-203	44	33	return	return	VERB
fbem-203	44	34	a	a	DET
fbem-203	44	35	lower	low	ADJ
fbem-203	44	36	similarity	similarity	NOUN
fbem-203	44	37	score	score	NOUN
fbem-203	44	38	.	.	PUNCT
fbem-203	45	1	to	to	PART
fbem-203	45	2	find	find	VERB
fbem-203	45	3	the	the	DET
fbem-203	45	4	new	new	ADJ
fbem-203	45	5	position	position	NOUN
fbem-203	45	6	of	of	ADP
fbem-203	45	7	the	the	DET
fbem-203	45	8	target	target	NOUN
fbem-203	45	9	in	in	ADP
fbem-203	45	10	the	the	DET
fbem-203	45	11	new	new	ADJ
fbem-203	45	12	video	video	NOUN
fbem-203	45	13	frame	frame	NOUN
fbem-203	45	14	image	image	NOUN
fbem-203	45	15	,	,	PUNCT
fbem-203	45	16	we	we	PRON
fbem-203	45	17	have	have	VERB
fbem-203	45	18	to	to	PART
fbem-203	45	19	perform	perform	VERB
fbem-203	45	20	a	a	DET
fbem-203	45	21	complete	complete	ADJ
fbem-203	45	22	test	test	NOUN
fbem-203	45	23	on	on	ADP
fbem-203	45	24	all	all	DET
fbem-203	45	25	possible	possible	ADJ
fbem-203	45	26	positions	position	NOUN
fbem-203	45	27	and	and	CCONJ
fbem-203	45	28	select	select	VERB
fbem-203	45	29	the	the	DET
fbem-203	45	30	candidate	candidate	NOUN
fbem-203	45	31	target	target	NOUN
fbem-203	45	32	with	with	ADP
fbem-203	45	33	the	the	DET
fbem-203	45	34	highest	high	ADJ
fbem-203	45	35	similarity	similarity	NOUN
fbem-203	45	36	.	.	PUNCT
fbem-203	46	1	the	the	DET
fbem-203	46	2	mapping	mapping	NOUN
fbem-203	46	3	function	function	NOUN
fbem-203	46	4	f(z	f(z	PROPN
fbem-203	46	5	,	,	PUNCT
fbem-203	46	6	x	x	X
fbem-203	46	7	)	)	PUNCT
fbem-203	46	8	is	be	AUX
fbem-203	46	9	shown	show	VERB
fbem-203	46	10	in	in	ADP
fbem-203	46	11	the	the	DET
fbem-203	46	12	following	follow	VERB
fbem-203	46	13	formula	formula	NOUN
fbem-203	46	14	1	1	NUM
fbem-203	46	15	,	,	PUNCT
fbem-203	46	16	where	where	SCONJ
fbem-203	46	17	b_1	b_1	PROPN
fbem-203	46	18	represents	represent	VERB
fbem-203	46	19	the	the	DET
fbem-203	46	20	offset	offset	ADJ
fbem-203	46	21	signal	signal	NOUN
fbem-203	46	22	of	of	ADP
fbem-203	46	23	the	the	DET
fbem-203	46	24	position	position	NOUN
fbem-203	46	25	fetching	fetch	VERB
fbem-203	46	26	point	point	NOUN
fbem-203	46	27	.	.	PUNCT
fbem-203	47	1	79	79	NUM
fbem-203	48	1	f	f	X
fbem-203	48	2	z	z	NOUN
fbem-203	48	3	,	,	PUNCT
fbem-203	48	4	x	x	PROPN
fbem-203	48	5	g	g	PROPN
fbem-203	48	6	φ	φ	PROPN
fbem-203	48	7	z	z	PROPN
fbem-203	48	8	,	,	PUNCT
fbem-203	48	9	x	x	PUNCT
fbem-203	49	1	=	=	NOUN
fbem-203	49	2	φ	φ	PROPN
fbem-203	49	3	∗	∗	PROPN
fbem-203	49	4	φ	φ	PROPN
fbem-203	49	5	b	b	PROPN
fbem-203	49	6	(	(	PUNCT
fbem-203	49	7	1	1	NUM
fbem-203	49	8	)	)	PUNCT
fbem-203	49	9	the	the	DET
fbem-203	49	10	structure	structure	NOUN
fbem-203	49	11	diagram	diagram	NOUN
fbem-203	49	12	of	of	ADP
fbem-203	49	13	the	the	DET
fbem-203	49	14	tracking	tracking	NOUN
fbem-203	49	15	algorithm	algorithm	NOUN
fbem-203	49	16	based	base	VERB
fbem-203	49	17	on	on	ADP
fbem-203	49	18	the	the	DET
fbem-203	49	19	twin	twin	ADJ
fbem-203	49	20	neural	neural	ADJ
fbem-203	49	21	network	network	NOUN
fbem-203	49	22	is	be	AUX
fbem-203	49	23	shown	show	VERB
fbem-203	49	24	in	in	ADP
fbem-203	49	25	figure	figure	NOUN
fbem-203	49	26	1	1	NUM
fbem-203	49	27	below	below	ADV
fbem-203	49	28	:	:	PUNCT
fbem-203	49	29	z	z	NOUN
fbem-203	49	30	x	x	PUNCT
fbem-203	49	31	ψ	ψ	X
fbem-203	49	32	ψ	ψ	X
fbem-203	49	33	*	*	PUNCT
fbem-203	49	34	127	127	NUM
fbem-203	49	35	*	*	SYM
fbem-203	49	36	127	127	NUM
fbem-203	49	37	*	*	SYM
fbem-203	49	38	3	3	NUM
fbem-203	49	39	255	255	NUM
fbem-203	49	40	*	*	NUM
fbem-203	49	41	255	255	NUM
fbem-203	49	42	*	*	SYM
fbem-203	49	43	3	3	NUM
fbem-203	49	44	22	22	NUM
fbem-203	49	45	*	*	SYM
fbem-203	49	46	22	22	NUM
fbem-203	49	47	*	*	SYM
fbem-203	49	48	128	128	NUM
fbem-203	49	49	6	6	NUM
fbem-203	49	50	*	*	SYM
fbem-203	49	51	6	6	NUM
fbem-203	49	52	*	*	SYM
fbem-203	49	53	128	128	NUM
fbem-203	49	54	17	17	NUM
fbem-203	49	55	*	*	SYM
fbem-203	49	56	17	17	NUM
fbem-203	49	57	*	*	SYM
fbem-203	49	58	1	1	NUM
fbem-203	49	59	figure	figure	NOUN
fbem-203	49	60	1	1	NUM
fbem-203	49	61	.	.	PUNCT
fbem-203	49	62	structural	structural	ADJ
fbem-203	49	63	diagram	diagram	NOUN
fbem-203	49	64	of	of	ADP
fbem-203	49	65	tracking	track	VERB
fbem-203	49	66	algorithm	algorithm	NOUN
fbem-203	49	67	based	base	VERB
fbem-203	49	68	on	on	ADP
fbem-203	49	69	siamese	siamese	ADJ
fbem-203	49	70	network	network	NOUN
fbem-203	49	71	as	as	SCONJ
fbem-203	49	72	shown	show	VERB
fbem-203	49	73	in	in	ADP
fbem-203	49	74	figure	figure	NOUN
fbem-203	49	75	1	1	NUM
fbem-203	49	76	,	,	PUNCT
fbem-203	49	77	z	z	PROPN
fbem-203	49	78	represents	represent	VERB
fbem-203	49	79	the	the	DET
fbem-203	49	80	template	template	NOUN
fbem-203	49	81	image	image	NOUN
fbem-203	49	82	(	(	PUNCT
fbem-203	49	83	the	the	DET
fbem-203	49	84	real	real	ADJ
fbem-203	49	85	target	target	NOUN
fbem-203	49	86	in	in	ADP
fbem-203	49	87	the	the	DET
fbem-203	49	88	first	first	ADJ
fbem-203	49	89	frame	frame	NOUN
fbem-203	49	90	)	)	PUNCT
fbem-203	49	91	,	,	PUNCT
fbem-203	49	92	x	x	PRON
fbem-203	49	93	represents	represent	VERB
fbem-203	49	94	the	the	DET
fbem-203	49	95	search	search	NOUN
fbem-203	49	96	image	image	NOUN
fbem-203	49	97	(	(	PUNCT
fbem-203	49	98	the	the	DET
fbem-203	49	99	search	search	NOUN
fbem-203	49	100	area	area	NOUN
fbem-203	49	101	acquired	acquire	VERB
fbem-203	49	102	at	at	ADP
fbem-203	49	103	the	the	DET
fbem-203	49	104	center	center	NOUN
fbem-203	49	105	is	be	AUX
fbem-203	49	106	the	the	DET
fbem-203	49	107	target	target	NOUN
fbem-203	49	108	center	center	NOUN
fbem-203	49	109	of	of	ADP
fbem-203	49	110	the	the	DET
fbem-203	49	111	previous	previous	ADJ
fbem-203	49	112	frame	frame	NOUN
fbem-203	49	113	)	)	PUNCT
fbem-203	49	114	,	,	PUNCT
fbem-203	49	115	and	and	CCONJ
fbem-203	49	116	the	the	DET
fbem-203	49	117	template	template	NOUN
fbem-203	49	118	image	image	NOUN
fbem-203	49	119	and	and	CCONJ
fbem-203	49	120	the	the	DET
fbem-203	49	121	search	search	NOUN
fbem-203	49	122	image	image	NOUN
fbem-203	49	123	are	be	AUX
fbem-203	49	124	the	the	DET
fbem-203	49	125	input	input	NOUN
fbem-203	49	126	network	network	NOUN
fbem-203	49	127	.	.	PUNCT
fbem-203	50	1	φ	φ	PROPN
fbem-203	50	2	refers	refer	VERB
fbem-203	50	3	to	to	ADP
fbem-203	50	4	the	the	DET
fbem-203	50	5	depth	depth	NOUN
fbem-203	50	6	features	feature	NOUN
fbem-203	50	7	extracted	extract	VERB
fbem-203	50	8	from	from	ADP
fbem-203	50	9	the	the	DET
fbem-203	50	10	corresponding	corresponding	ADJ
fbem-203	50	11	images	image	NOUN
fbem-203	50	12	of	of	ADP
fbem-203	50	13	each	each	DET
fbem-203	50	14	branch	branch	NOUN
fbem-203	50	15	,	,	PUNCT
fbem-203	50	16	and	and	CCONJ
fbem-203	50	17	*	*	PUNCT
fbem-203	50	18	represents	represent	VERB
fbem-203	50	19	the	the	DET
fbem-203	50	20	cross	cross	ADJ
fbem-203	50	21	-	-	ADJ
fbem-203	50	22	correlation	correlation	ADJ
fbem-203	50	23	operation	operation	NOUN
fbem-203	50	24	.	.	PUNCT
fbem-203	51	1	2.2	2.2	NUM
fbem-203	51	2	.	.	PUNCT
fbem-203	51	3	attention	attention	NOUN
fbem-203	51	4	mechanism	mechanism	NOUN
fbem-203	51	5	in	in	ADP
fbem-203	51	6	the	the	DET
fbem-203	51	7	field	field	NOUN
fbem-203	51	8	of	of	ADP
fbem-203	51	9	image	image	NOUN
fbem-203	51	10	processing	processing	NOUN
fbem-203	51	11	,	,	PUNCT
fbem-203	51	12	the	the	DET
fbem-203	51	13	attention	attention	NOUN
fbem-203	51	14	mechanism	mechanism	NOUN
fbem-203	51	15	draws	draw	VERB
fbem-203	51	16	on	on	ADP
fbem-203	51	17	the	the	DET
fbem-203	51	18	human	human	ADJ
fbem-203	51	19	visual	visual	ADJ
fbem-203	51	20	mechanism	mechanism	NOUN
fbem-203	51	21	,	,	PUNCT
fbem-203	51	22	while	while	SCONJ
fbem-203	51	23	the	the	DET
fbem-203	51	24	network	network	NOUN
fbem-203	51	25	focus	focus	NOUN
fbem-203	51	26	is	be	AUX
fbem-203	51	27	limited	limit	VERB
fbem-203	51	28	to	to	ADP
fbem-203	51	29	a	a	DET
fbem-203	51	30	sub	sub	NOUN
fbem-203	51	31	-	-	NOUN
fbem-203	51	32	region	region	NOUN
fbem-203	51	33	of	of	ADP
fbem-203	51	34	the	the	DET
fbem-203	51	35	image	image	NOUN
fbem-203	51	36	.	.	PUNCT
fbem-203	52	1	the	the	DET
fbem-203	52	2	emphasis	emphasis	NOUN
fbem-203	52	3	on	on	ADP
fbem-203	52	4	the	the	DET
fbem-203	52	5	target	target	NOUN
fbem-203	52	6	region	region	NOUN
fbem-203	52	7	improves	improve	VERB
fbem-203	52	8	the	the	DET
fbem-203	52	9	feature	feature	NOUN
fbem-203	52	10	expression	expression	NOUN
fbem-203	52	11	ability	ability	NOUN
fbem-203	52	12	of	of	ADP
fbem-203	52	13	the	the	DET
fbem-203	52	14	model	model	NOUN
fbem-203	52	15	.	.	PUNCT
fbem-203	53	1	it	it	PRON
fbem-203	53	2	is	be	AUX
fbem-203	53	3	similar	similar	ADJ
fbem-203	53	4	to	to	ADP
fbem-203	53	5	the	the	DET
fbem-203	53	6	human	human	ADJ
fbem-203	53	7	cognition	cognition	NOUN
fbem-203	53	8	process	process	NOUN
fbem-203	53	9	,	,	PUNCT
fbem-203	53	10	which	which	PRON
fbem-203	53	11	can	can	AUX
fbem-203	53	12	selectively	selectively	ADV
fbem-203	53	13	focus	focus	VERB
fbem-203	53	14	on	on	ADP
fbem-203	53	15	part	part	NOUN
fbem-203	53	16	of	of	ADP
fbem-203	53	17	all	all	DET
fbem-203	53	18	information	information	NOUN
fbem-203	53	19	while	while	SCONJ
fbem-203	53	20	ignoring	ignore	VERB
fbem-203	53	21	other	other	ADJ
fbem-203	53	22	information	information	NOUN
fbem-203	53	23	.	.	PUNCT
fbem-203	54	1	in	in	ADP
fbem-203	54	2	the	the	DET
fbem-203	54	3	neural	neural	ADJ
fbem-203	54	4	network	network	NOUN
fbem-203	54	5	structure	structure	NOUN
fbem-203	54	6	,	,	PUNCT
fbem-203	54	7	the	the	DET
fbem-203	54	8	attention	attention	NOUN
fbem-203	54	9	model	model	NOUN
fbem-203	54	10	achieves	achieve	VERB
fbem-203	54	11	attention	attention	NOUN
fbem-203	54	12	to	to	ADP
fbem-203	54	13	specific	specific	ADJ
fbem-203	54	14	information	information	NOUN
fbem-203	54	15	by	by	ADP
fbem-203	54	16	assigning	assign	VERB
fbem-203	54	17	network	network	NOUN
fbem-203	54	18	weights	weight	NOUN
fbem-203	54	19	.	.	PUNCT
fbem-203	55	1	in	in	ADP
fbem-203	55	2	recent	recent	ADJ
fbem-203	55	3	years	year	NOUN
fbem-203	55	4	,	,	PUNCT
fbem-203	55	5	with	with	ADP
fbem-203	55	6	the	the	DET
fbem-203	55	7	continuous	continuous	ADJ
fbem-203	55	8	research	research	NOUN
fbem-203	55	9	of	of	ADP
fbem-203	55	10	attention	attention	NOUN
fbem-203	55	11	model	model	NOUN
fbem-203	55	12	,	,	PUNCT
fbem-203	55	13	many	many	ADJ
fbem-203	55	14	effective	effective	ADJ
fbem-203	55	15	attention	attention	NOUN
fbem-203	55	16	model	model	NOUN
fbem-203	55	17	network	network	NOUN
fbem-203	55	18	structures	structure	NOUN
fbem-203	55	19	have	have	AUX
fbem-203	55	20	appeared	appear	VERB
fbem-203	55	21	.	.	PUNCT
fbem-203	56	1	the	the	DET
fbem-203	56	2	attention	attention	NOUN
fbem-203	56	3	model	model	NOUN
fbem-203	56	4	is	be	AUX
fbem-203	56	5	widely	widely	ADV
fbem-203	56	6	used	use	VERB
fbem-203	56	7	in	in	ADP
fbem-203	56	8	many	many	ADJ
fbem-203	56	9	aspects	aspect	NOUN
fbem-203	56	10	such	such	ADJ
fbem-203	56	11	as	as	ADP
fbem-203	56	12	target	target	NOUN
fbem-203	56	13	detection	detection	NOUN
fbem-203	56	14	,	,	PUNCT
fbem-203	56	15	target	target	VERB
fbem-203	56	16	classification	classification	NOUN
fbem-203	56	17	,	,	PUNCT
fbem-203	56	18	and	and	CCONJ
fbem-203	56	19	target	target	VERB
fbem-203	56	20	tracking	tracking	NOUN
fbem-203	56	21	.	.	PUNCT
fbem-203	57	1	training	training	NOUN
fbem-203	57	2	can	can	AUX
fbem-203	57	3	improve	improve	VERB
fbem-203	57	4	the	the	DET
fbem-203	57	5	model	model	NOUN
fbem-203	57	6	's	's	PART
fbem-203	57	7	ability	ability	NOUN
fbem-203	57	8	to	to	PART
fbem-203	57	9	model	model	VERB
fbem-203	57	10	the	the	DET
fbem-203	57	11	space	space	NOUN
fbem-203	57	12	,	,	PUNCT
fbem-203	57	13	channel	channel	NOUN
fbem-203	57	14	,	,	PUNCT
fbem-203	57	15	background	background	NOUN
fbem-203	57	16	and	and	CCONJ
fbem-203	57	17	other	other	ADJ
fbem-203	57	18	information	information	NOUN
fbem-203	57	19	of	of	ADP
fbem-203	57	20	the	the	DET
fbem-203	57	21	neural	neural	ADJ
fbem-203	57	22	network	network	NOUN
fbem-203	57	23	,	,	PUNCT
fbem-203	57	24	which	which	PRON
fbem-203	57	25	helps	help	VERB
fbem-203	57	26	to	to	PART
fbem-203	57	27	improve	improve	VERB
fbem-203	57	28	the	the	DET
fbem-203	57	29	representation	representation	NOUN
fbem-203	57	30	performance	performance	NOUN
fbem-203	57	31	of	of	ADP
fbem-203	57	32	the	the	DET
fbem-203	57	33	convolutional	convolutional	ADJ
fbem-203	57	34	neural	neural	ADJ
fbem-203	57	35	network	network	NOUN
fbem-203	57	36	2.3	2.3	NUM
fbem-203	57	37	.	.	PUNCT
fbem-203	58	1	spatial	spatial	ADJ
fbem-203	58	2	attention	attention	NOUN
fbem-203	58	3	mechanism	mechanism	NOUN
fbem-203	58	4	each	each	DET
fbem-203	58	5	picture	picture	NOUN
fbem-203	58	6	contains	contain	VERB
fbem-203	58	7	a	a	DET
fbem-203	58	8	wealth	wealth	NOUN
fbem-203	58	9	of	of	ADP
fbem-203	58	10	data	datum	NOUN
fbem-203	58	11	information	information	NOUN
fbem-203	58	12	,	,	PUNCT
fbem-203	58	13	and	and	CCONJ
fbem-203	58	14	pixels	pixel	NOUN
fbem-203	58	15	and	and	CCONJ
fbem-203	58	16	pixels	pixel	NOUN
fbem-203	58	17	contain	contain	VERB
fbem-203	58	18	a	a	DET
fbem-203	58	19	certain	certain	ADJ
fbem-203	58	20	special	special	ADJ
fbem-203	58	21	connection	connection	NOUN
fbem-203	58	22	,	,	PUNCT
fbem-203	58	23	which	which	PRON
fbem-203	58	24	can	can	AUX
fbem-203	58	25	be	be	AUX
fbem-203	58	26	used	use	VERB
fbem-203	58	27	to	to	PART
fbem-203	58	28	obtain	obtain	VERB
fbem-203	58	29	features	feature	NOUN
fbem-203	58	30	with	with	ADP
fbem-203	58	31	more	more	ADJ
fbem-203	58	32	semantic	semantic	ADJ
fbem-203	58	33	information	information	NOUN
fbem-203	58	34	.	.	PUNCT
fbem-203	59	1	this	this	DET
fbem-203	59	2	paper	paper	NOUN
fbem-203	59	3	introduces	introduce	VERB
fbem-203	59	4	the	the	DET
fbem-203	59	5	spatial	spatial	ADJ
fbem-203	59	6	attention	attention	NOUN
fbem-203	59	7	mechanism	mechanism	NOUN
fbem-203	59	8	,	,	PUNCT
fbem-203	59	9	which	which	PRON
fbem-203	59	10	is	be	AUX
fbem-203	59	11	to	to	PART
fbem-203	59	12	give	give	VERB
fbem-203	59	13	different	different	ADJ
fbem-203	59	14	weights	weight	NOUN
fbem-203	59	15	to	to	ADP
fbem-203	59	16	different	different	ADJ
fbem-203	59	17	positions	position	NOUN
fbem-203	59	18	in	in	ADP
fbem-203	59	19	different	different	ADJ
fbem-203	59	20	spatial	spatial	ADJ
fbem-203	59	21	positions	position	NOUN
fbem-203	59	22	during	during	ADP
fbem-203	59	23	feature	feature	NOUN
fbem-203	59	24	extraction	extraction	NOUN
fbem-203	59	25	,	,	PUNCT
fbem-203	59	26	so	so	SCONJ
fbem-203	59	27	as	as	SCONJ
fbem-203	59	28	to	to	PART
fbem-203	59	29	achieve	achieve	VERB
fbem-203	59	30	the	the	DET
fbem-203	59	31	extraction	extraction	NOUN
fbem-203	59	32	of	of	ADP
fbem-203	59	33	different	different	ADJ
fbem-203	59	34	feature	feature	NOUN
fbem-203	59	35	positions	position	NOUN
fbem-203	59	36	on	on	ADP
fbem-203	59	37	the	the	DET
fbem-203	59	38	feature	feature	NOUN
fbem-203	59	39	map	map	NOUN
fbem-203	59	40	,	,	PUNCT
fbem-203	59	41	taking	take	VERB
fbem-203	59	42	the	the	DET
fbem-203	59	43	same	same	ADJ
fbem-203	59	44	position	position	NOUN
fbem-203	59	45	between	between	ADP
fbem-203	59	46	the	the	DET
fbem-203	59	47	feature	feature	NOUN
fbem-203	59	48	maps	map	NOUN
fbem-203	59	49	as	as	ADP
fbem-203	59	50	the	the	DET
fbem-203	59	51	unit	unit	NOUN
fbem-203	59	52	,	,	PUNCT
fbem-203	59	53	by	by	ADP
fbem-203	59	54	comparing	compare	VERB
fbem-203	59	55	the	the	DET
fbem-203	59	56	similarity	similarity	NOUN
fbem-203	59	57	between	between	ADP
fbem-203	59	58	the	the	DET
fbem-203	59	59	feature	feature	NOUN
fbem-203	59	60	maps	map	NOUN
fbem-203	59	61	,	,	PUNCT
fbem-203	59	62	pays	pay	VERB
fbem-203	59	63	attention	attention	NOUN
fbem-203	59	64	to	to	ADP
fbem-203	59	65	the	the	DET
fbem-203	59	66	importance	importance	NOUN
fbem-203	59	67	of	of	ADP
fbem-203	59	68	a	a	DET
fbem-203	59	69	certain	certain	ADJ
fbem-203	59	70	position	position	NOUN
fbem-203	59	71	in	in	ADP
fbem-203	59	72	the	the	DET
fbem-203	59	73	spatial	spatial	ADJ
fbem-203	59	74	position	position	NOUN
fbem-203	59	75	.	.	PUNCT
fbem-203	60	1	for	for	ADP
fbem-203	60	2	each	each	DET
fbem-203	60	3	feature	feature	NOUN
fbem-203	60	4	in	in	ADP
fbem-203	60	5	the	the	DET
fbem-203	60	6	same	same	ADJ
fbem-203	60	7	position	position	NOUN
fbem-203	60	8	in	in	ADP
fbem-203	60	9	the	the	DET
fbem-203	60	10	two	two	NUM
fbem-203	60	11	feature	feature	NOUN
fbem-203	60	12	maps	map	NOUN
fbem-203	60	13	,	,	PUNCT
fbem-203	60	14	it	it	PRON
fbem-203	60	15	can	can	AUX
fbem-203	60	16	be	be	AUX
fbem-203	60	17	updated	update	VERB
fbem-203	60	18	by	by	ADP
fbem-203	60	19	the	the	DET
fbem-203	60	20	weighted	weight	VERB
fbem-203	60	21	sum	sum	NOUN
fbem-203	60	22	method	method	NOUN
fbem-203	60	23	to	to	PART
fbem-203	60	24	strengthen	strengthen	VERB
fbem-203	60	25	or	or	CCONJ
fbem-203	60	26	suppress	suppress	VERB
fbem-203	60	27	the	the	DET
fbem-203	60	28	weights	weight	NOUN
fbem-203	60	29	of	of	ADP
fbem-203	60	30	different	different	ADJ
fbem-203	60	31	spatial	spatial	ADJ
fbem-203	60	32	location	location	NOUN
fbem-203	60	33	features	feature	NOUN
fbem-203	60	34	,	,	PUNCT
fbem-203	60	35	the	the	PRON
fbem-203	60	36	greater	great	ADJ
fbem-203	60	37	the	the	DET
fbem-203	60	38	weight	weight	NOUN
fbem-203	60	39	,	,	PUNCT
fbem-203	60	40	the	the	PRON
fbem-203	60	41	more	more	ADV
fbem-203	60	42	similar	similar	ADJ
fbem-203	60	43	the	the	DET
fbem-203	60	44	features	feature	NOUN
fbem-203	60	45	.	.	PUNCT
fbem-203	61	1	by	by	ADP
fbem-203	61	2	introducing	introduce	VERB
fbem-203	61	3	the	the	DET
fbem-203	61	4	spatial	spatial	ADJ
fbem-203	61	5	attention	attention	NOUN
fbem-203	61	6	mechanism	mechanism	NOUN
fbem-203	61	7	,	,	PUNCT
fbem-203	61	8	it	it	PRON
fbem-203	61	9	is	be	AUX
fbem-203	61	10	possible	possible	ADJ
fbem-203	61	11	to	to	PART
fbem-203	61	12	increase	increase	VERB
fbem-203	61	13	the	the	DET
fbem-203	61	14	spatial	spatial	ADJ
fbem-203	61	15	weight	weight	NOUN
fbem-203	61	16	of	of	ADP
fbem-203	61	17	important	important	ADJ
fbem-203	61	18	features	feature	NOUN
fbem-203	61	19	,	,	PUNCT
fbem-203	61	20	effectively	effectively	ADV
fbem-203	61	21	extract	extract	VERB
fbem-203	61	22	important	important	ADJ
fbem-203	61	23	features	feature	NOUN
fbem-203	61	24	,	,	PUNCT
fbem-203	61	25	and	and	CCONJ
fbem-203	61	26	effectively	effectively	ADV
fbem-203	61	27	improve	improve	VERB
fbem-203	61	28	the	the	DET
fbem-203	61	29	feature	feature	NOUN
fbem-203	61	30	expression	expression	NOUN
fbem-203	61	31	ability	ability	NOUN
fbem-203	61	32	of	of	ADP
fbem-203	61	33	the	the	DET
fbem-203	61	34	algorithm	algorithm	NOUN
fbem-203	61	35	model	model	NOUN
fbem-203	61	36	without	without	ADP
fbem-203	61	37	affecting	affect	VERB
fbem-203	61	38	the	the	DET
fbem-203	61	39	computational	computational	ADJ
fbem-203	61	40	complexity	complexity	NOUN
fbem-203	61	41	and	and	CCONJ
fbem-203	61	42	speed	speed	NOUN
fbem-203	61	43	of	of	ADP
fbem-203	61	44	the	the	DET
fbem-203	61	45	algorithm	algorithm	NOUN
fbem-203	61	46	.	.	PUNCT
fbem-203	62	1	figure	figure	NOUN
fbem-203	62	2	2	2	NUM
fbem-203	62	3	shows	show	VERB
fbem-203	62	4	the	the	DET
fbem-203	62	5	structure	structure	NOUN
fbem-203	62	6	of	of	ADP
fbem-203	62	7	this	this	DET
fbem-203	62	8	module	module	NOUN
fbem-203	62	9	.	.	PUNCT
fbem-203	63	1	figure	figure	NOUN
fbem-203	63	2	2	2	NUM
fbem-203	63	3	.	.	PUNCT
fbem-203	63	4	spatial	spatial	ADJ
fbem-203	63	5	attention	attention	NOUN
fbem-203	63	6	module	module	NOUN
fbem-203	63	7	2.4	2.4	NUM
fbem-203	63	8	.	.	PUNCT
fbem-203	64	1	residual	residual	ADJ
fbem-203	64	2	network	network	NOUN
fbem-203	64	3	the	the	DET
fbem-203	64	4	increase	increase	NOUN
fbem-203	64	5	in	in	ADP
fbem-203	64	6	model	model	NOUN
fbem-203	64	7	depth	depth	NOUN
fbem-203	64	8	in	in	ADP
fbem-203	64	9	the	the	DET
fbem-203	64	10	network	network	NOUN
fbem-203	64	11	caused	cause	VERB
fbem-203	64	12	a	a	DET
fbem-203	64	13	series	series	NOUN
fbem-203	64	14	of	of	ADP
fbem-203	64	15	gradient	gradient	NOUN
fbem-203	64	16	-	-	PUNCT
fbem-203	64	17	related	relate	VERB
fbem-203	64	18	problems	problem	NOUN
fbem-203	64	19	.	.	PUNCT
fbem-203	65	1	scholars	scholar	NOUN
fbem-203	65	2	found	find	VERB
fbem-203	65	3	in	in	ADP
fbem-203	65	4	the	the	DET
fbem-203	65	5	development	development	NOUN
fbem-203	65	6	of	of	ADP
fbem-203	65	7	convolutional	convolutional	ADJ
fbem-203	65	8	neural	neural	ADJ
fbem-203	65	9	networks	network	NOUN
fbem-203	65	10	that	that	PRON
fbem-203	65	11	the	the	DET
fbem-203	65	12	performance	performance	NOUN
fbem-203	65	13	of	of	ADP
fbem-203	65	14	neural	neural	ADJ
fbem-203	65	15	networks	network	NOUN
fbem-203	65	16	is	be	AUX
fbem-203	65	17	affected	affect	VERB
fbem-203	65	18	by	by	ADP
fbem-203	65	19	the	the	DET
fbem-203	65	20	depth	depth	NOUN
fbem-203	65	21	of	of	ADP
fbem-203	65	22	the	the	DET
fbem-203	65	23	network	network	NOUN
fbem-203	65	24	.	.	PUNCT
fbem-203	66	1	as	as	SCONJ
fbem-203	66	2	the	the	DET
fbem-203	66	3	number	number	NOUN
fbem-203	66	4	of	of	ADP
fbem-203	66	5	network	network	NOUN
fbem-203	66	6	layers	layer	NOUN
fbem-203	66	7	increases	increase	VERB
fbem-203	66	8	,	,	PUNCT
fbem-203	66	9	the	the	PRON
fbem-203	66	10	better	well	ADJ
fbem-203	66	11	the	the	DET
fbem-203	66	12	model	model	NOUN
fbem-203	66	13	performance	performance	NOUN
fbem-203	66	14	,	,	PUNCT
fbem-203	66	15	but	but	CCONJ
fbem-203	66	16	also	also	ADV
fbem-203	66	17	the	the	DET
fbem-203	66	18	problem	problem	NOUN
fbem-203	66	19	of	of	ADP
fbem-203	66	20	gradient	gradient	ADJ
fbem-203	66	21	disappearance	disappearance	NOUN
fbem-203	66	22	and	and	CCONJ
fbem-203	66	23	explosion	explosion	NOUN
fbem-203	66	24	will	will	AUX
fbem-203	66	25	eventually	eventually	ADV
fbem-203	66	26	hinder	hinder	VERB
fbem-203	66	27	the	the	DET
fbem-203	66	28	convergence	convergence	NOUN
fbem-203	66	29	of	of	ADP
fbem-203	66	30	the	the	DET
fbem-203	66	31	model	model	NOUN
fbem-203	66	32	due	due	ADP
fbem-203	66	33	to	to	ADP
fbem-203	66	34	the	the	DET
fbem-203	66	35	failure	failure	NOUN
fbem-203	66	36	of	of	ADP
fbem-203	66	37	network	network	NOUN
fbem-203	66	38	parameter	parameter	NOUN
fbem-203	66	39	update	update	NOUN
fbem-203	66	40	.	.	PUNCT
fbem-203	67	1	in	in	ADP
fbem-203	67	2	order	order	NOUN
fbem-203	67	3	to	to	PART
fbem-203	67	4	solve	solve	VERB
fbem-203	67	5	this	this	DET
fbem-203	67	6	problem	problem	NOUN
fbem-203	67	7	,	,	PUNCT
fbem-203	67	8	he	he	PRON
fbem-203	67	9	et	et	PROPN
fbem-203	67	10	al	al	PROPN
fbem-203	67	11	.	.	PROPN
fbem-203	67	12	proposed	propose	VERB
fbem-203	67	13	a	a	DET
fbem-203	67	14	convolutional	convolutional	ADJ
fbem-203	67	15	neural	neural	ADJ
fbem-203	67	16	network	network	NOUN
fbem-203	67	17	called	call	VERB
fbem-203	67	18	residual	residual	ADJ
fbem-203	67	19	net	net	NOUN
fbem-203	67	20	in	in	ADP
fbem-203	67	21	2015	2015	NUM
fbem-203	67	22	,	,	PUNCT
fbem-203	67	23	which	which	PRON
fbem-203	67	24	trains	train	VERB
fbem-203	67	25	cnn	cnn	PROPN
fbem-203	67	26	to	to	ADP
fbem-203	67	27	a	a	DET
fbem-203	67	28	depth	depth	NOUN
fbem-203	67	29	of	of	ADP
fbem-203	67	30	152	152	NUM
fbem-203	67	31	layers	layer	NOUN
fbem-203	67	32	under	under	ADP
fbem-203	67	33	the	the	DET
fbem-203	67	34	action	action	NOUN
fbem-203	67	35	of	of	ADP
fbem-203	67	36	the	the	DET
fbem-203	67	37	internal	internal	ADJ
fbem-203	67	38	residual	residual	ADJ
fbem-203	67	39	module	module	NOUN
fbem-203	67	40	,	,	PUNCT
fbem-203	67	41	and	and	CCONJ
fbem-203	67	42	performs	perform	VERB
fbem-203	67	43	multiple	multiple	ADJ
fbem-203	67	44	detection	detection	NOUN
fbem-203	67	45	,	,	PUNCT
fbem-203	67	46	segmentation	segmentation	NOUN
fbem-203	67	47	,	,	PUNCT
fbem-203	67	48	and	and	CCONJ
fbem-203	67	49	first	first	ADJ
fbem-203	67	50	place	place	NOUN
fbem-203	67	51	in	in	ADP
fbem-203	67	52	the	the	DET
fbem-203	67	53	positioning	positioning	NOUN
fbem-203	67	54	competition	competition	NOUN
fbem-203	67	55	.	.	PUNCT
fbem-203	68	1	the	the	DET
fbem-203	68	2	network	network	NOUN
fbem-203	68	3	introduces	introduce	VERB
fbem-203	68	4	the	the	DET
fbem-203	68	5	input	input	NOUN
fbem-203	68	6	characteristics	characteristic	NOUN
fbem-203	68	7	to	to	ADP
fbem-203	68	8	the	the	DET
fbem-203	68	9	output	output	NOUN
fbem-203	68	10	,	,	PUNCT
fbem-203	68	11	does	do	AUX
fbem-203	68	12	not	not	PART
fbem-203	68	13	increase	increase	VERB
fbem-203	68	14	the	the	DET
fbem-203	68	15	complexity	complexity	NOUN
fbem-203	68	16	of	of	ADP
fbem-203	68	17	the	the	DET
fbem-203	68	18	network	network	NOUN
fbem-203	68	19	model	model	NOUN
fbem-203	68	20	,	,	PUNCT
fbem-203	68	21	and	and	CCONJ
fbem-203	68	22	reduces	reduce	VERB
fbem-203	68	23	the	the	DET
fbem-203	68	24	error	error	NOUN
fbem-203	68	25	rate	rate	NOUN
fbem-203	68	26	at	at	ADP
fbem-203	68	27	the	the	DET
fbem-203	68	28	same	same	ADJ
fbem-203	68	29	time	time	NOUN
fbem-203	68	30	.	.	PUNCT
fbem-203	69	1	80	80	NUM
fbem-203	69	2	3	3	NUM
fbem-203	69	3	.	.	PUNCT
fbem-203	70	1	the	the	DET
fbem-203	70	2	algorithm	algorithm	NOUN
fbem-203	70	3	of	of	ADP
fbem-203	70	4	this	this	DET
fbem-203	70	5	paper	paper	NOUN
fbem-203	70	6	the	the	DET
fbem-203	70	7	siamfc	siamfc	NOUN
fbem-203	70	8	algorithm	algorithm	NOUN
fbem-203	70	9	(	(	PUNCT
fbem-203	70	10	siamfc	siamfc	NOUN
fbem-203	70	11	)	)	PUNCT
fbem-203	70	12	transforms	transform	VERB
fbem-203	70	13	the	the	DET
fbem-203	70	14	tracking	tracking	NOUN
fbem-203	70	15	problem	problem	NOUN
fbem-203	70	16	into	into	ADP
fbem-203	70	17	solving	solve	VERB
fbem-203	70	18	the	the	DET
fbem-203	70	19	similarity	similarity	NOUN
fbem-203	70	20	problem	problem	NOUN
fbem-203	70	21	in	in	ADP
fbem-203	70	22	the	the	DET
fbem-203	70	23	target	target	NOUN
fbem-203	70	24	tracking	tracking	NOUN
fbem-203	70	25	process	process	NOUN
fbem-203	70	26	.	.	PUNCT
fbem-203	71	1	although	although	SCONJ
fbem-203	71	2	the	the	DET
fbem-203	71	3	algorithm	algorithm	NOUN
fbem-203	71	4	has	have	AUX
fbem-203	71	5	achieved	achieve	VERB
fbem-203	71	6	good	good	ADJ
fbem-203	71	7	results	result	NOUN
fbem-203	71	8	in	in	ADP
fbem-203	71	9	the	the	DET
fbem-203	71	10	tracking	tracking	NOUN
fbem-203	71	11	process	process	NOUN
fbem-203	71	12	,	,	PUNCT
fbem-203	71	13	when	when	SCONJ
fbem-203	71	14	the	the	DET
fbem-203	71	15	background	background	NOUN
fbem-203	71	16	changes	change	VERB
fbem-203	71	17	many	many	ADJ
fbem-203	71	18	times	time	NOUN
fbem-203	71	19	during	during	ADP
fbem-203	71	20	the	the	DET
fbem-203	71	21	movement	movement	NOUN
fbem-203	71	22	,	,	PUNCT
fbem-203	71	23	the	the	DET
fbem-203	71	24	algorithm	algorithm	NOUN
fbem-203	71	25	it	it	PRON
fbem-203	71	26	may	may	AUX
fbem-203	71	27	be	be	AUX
fbem-203	71	28	tracked	track	VERB
fbem-203	71	29	to	to	ADP
fbem-203	71	30	a	a	DET
fbem-203	71	31	target	target	NOUN
fbem-203	71	32	similar	similar	ADJ
fbem-203	71	33	to	to	ADP
fbem-203	71	34	the	the	DET
fbem-203	71	35	background	background	NOUN
fbem-203	71	36	in	in	ADP
fbem-203	71	37	the	the	DET
fbem-203	71	38	template	template	NOUN
fbem-203	71	39	,	,	PUNCT
fbem-203	71	40	resulting	result	VERB
fbem-203	71	41	in	in	ADP
fbem-203	71	42	tracking	track	VERB
fbem-203	71	43	drift	drift	NOUN
fbem-203	71	44	.	.	PUNCT
fbem-203	72	1	in	in	ADP
fbem-203	72	2	order	order	NOUN
fbem-203	72	3	to	to	PART
fbem-203	72	4	better	well	ADV
fbem-203	72	5	solve	solve	VERB
fbem-203	72	6	this	this	DET
fbem-203	72	7	problem	problem	NOUN
fbem-203	72	8	,	,	PUNCT
fbem-203	72	9	this	this	DET
fbem-203	72	10	paper	paper	NOUN
fbem-203	72	11	adopts	adopt	VERB
fbem-203	72	12	an	an	DET
fbem-203	72	13	adaptive	adaptive	ADJ
fbem-203	72	14	detailed	detailed	ADJ
fbem-203	72	15	feature	feature	NOUN
fbem-203	72	16	extraction	extraction	NOUN
fbem-203	72	17	,	,	PUNCT
fbem-203	72	18	adding	add	VERB
fbem-203	72	19	a	a	DET
fbem-203	72	20	residual	residual	ADJ
fbem-203	72	21	network	network	NOUN
fbem-203	72	22	to	to	ADP
fbem-203	72	23	the	the	DET
fbem-203	72	24	twin	twin	ADJ
fbem-203	72	25	network	network	NOUN
fbem-203	72	26	,	,	PUNCT
fbem-203	72	27	and	and	CCONJ
fbem-203	72	28	adding	add	VERB
fbem-203	72	29	it	it	PRON
fbem-203	72	30	to	to	ADP
fbem-203	72	31	each	each	DET
fbem-203	72	32	layer	layer	NOUN
fbem-203	72	33	they	they	PRON
fbem-203	72	34	are	be	AUX
fbem-203	72	35	all	all	ADV
fbem-203	72	36	embedded	embed	VERB
fbem-203	72	37	in	in	ADP
fbem-203	72	38	the	the	DET
fbem-203	72	39	detail	detail	NOUN
fbem-203	72	40	feature	feature	NOUN
fbem-203	72	41	retention	retention	NOUN
fbem-203	72	42	module	module	NOUN
fbem-203	72	43	,	,	PUNCT
fbem-203	72	44	amplify	amplify	VERB
fbem-203	72	45	the	the	DET
fbem-203	72	46	changes	change	NOUN
fbem-203	72	47	of	of	ADP
fbem-203	72	48	target	target	NOUN
fbem-203	72	49	features	feature	NOUN
fbem-203	72	50	,	,	PUNCT
fbem-203	72	51	retain	retain	VERB
fbem-203	72	52	the	the	DET
fbem-203	72	53	important	important	ADJ
fbem-203	72	54	structural	structural	ADJ
fbem-203	72	55	details	detail	NOUN
fbem-203	72	56	of	of	ADP
fbem-203	72	57	the	the	DET
fbem-203	72	58	original	original	ADJ
fbem-203	72	59	target	target	NOUN
fbem-203	72	60	features	feature	NOUN
fbem-203	72	61	,	,	PUNCT
fbem-203	72	62	play	play	VERB
fbem-203	72	63	a	a	DET
fbem-203	72	64	certain	certain	ADJ
fbem-203	72	65	role	role	NOUN
fbem-203	72	66	in	in	ADP
fbem-203	72	67	the	the	DET
fbem-203	72	68	search	search	NOUN
fbem-203	72	69	and	and	CCONJ
fbem-203	72	70	discrimination	discrimination	NOUN
fbem-203	72	71	of	of	ADP
fbem-203	72	72	details	detail	NOUN
fbem-203	72	73	in	in	ADP
fbem-203	72	74	target	target	NOUN
fbem-203	72	75	tracking	tracking	NOUN
fbem-203	72	76	,	,	PUNCT
fbem-203	72	77	and	and	CCONJ
fbem-203	72	78	solve	solve	VERB
fbem-203	72	79	the	the	DET
fbem-203	72	80	problem	problem	NOUN
fbem-203	72	81	of	of	ADP
fbem-203	72	82	the	the	DET
fbem-203	72	83	original	original	ADJ
fbem-203	72	84	tracking	tracking	NOUN
fbem-203	72	85	network	network	NOUN
fbem-203	72	86	when	when	SCONJ
fbem-203	72	87	the	the	DET
fbem-203	72	88	parameter	parameter	NOUN
fbem-203	72	89	scale	scale	NOUN
fbem-203	72	90	is	be	AUX
fbem-203	72	91	reduced	reduce	VERB
fbem-203	72	92	.	.	PUNCT
fbem-203	73	1	the	the	DET
fbem-203	73	2	problem	problem	NOUN
fbem-203	73	3	of	of	ADP
fbem-203	73	4	the	the	DET
fbem-203	73	5	loss	loss	NOUN
fbem-203	73	6	of	of	ADP
fbem-203	73	7	detailed	detailed	ADJ
fbem-203	73	8	features	feature	NOUN
fbem-203	73	9	,	,	PUNCT
fbem-203	73	10	and	and	CCONJ
fbem-203	73	11	the	the	DET
fbem-203	73	12	use	use	NOUN
fbem-203	73	13	of	of	ADP
fbem-203	73	14	a	a	DET
fbem-203	73	15	spatial	spatial	ADJ
fbem-203	73	16	attention	attention	NOUN
fbem-203	73	17	mechanism	mechanism	NOUN
fbem-203	73	18	,	,	PUNCT
fbem-203	73	19	is	be	AUX
fbem-203	73	20	used	use	VERB
fbem-203	73	21	to	to	PART
fbem-203	73	22	improve	improve	VERB
fbem-203	73	23	the	the	DET
fbem-203	73	24	network	network	NOUN
fbem-203	73	25	model	model	NOUN
fbem-203	73	26	's	's	PART
fbem-203	73	27	attention	attention	NOUN
fbem-203	73	28	to	to	ADP
fbem-203	73	29	effective	effective	ADJ
fbem-203	73	30	features	feature	NOUN
fbem-203	73	31	during	during	ADP
fbem-203	73	32	feature	feature	NOUN
fbem-203	73	33	extraction	extraction	NOUN
fbem-203	73	34	.	.	PUNCT
fbem-203	74	1	4	4	X
fbem-203	74	2	.	.	X
fbem-203	74	3	experiment	experiment	NOUN
fbem-203	74	4	and	and	CCONJ
fbem-203	74	5	analysis	analysis	NOUN
fbem-203	74	6	4.1	4.1	NUM
fbem-203	74	7	.	.	PUNCT
fbem-203	75	1	experimental	experimental	ADJ
fbem-203	75	2	environment	environment	NOUN
fbem-203	75	3	the	the	DET
fbem-203	75	4	algorithm	algorithm	NOUN
fbem-203	75	5	in	in	ADP
fbem-203	75	6	this	this	DET
fbem-203	75	7	paper	paper	NOUN
fbem-203	75	8	is	be	AUX
fbem-203	75	9	implemented	implement	VERB
fbem-203	75	10	based	base	VERB
fbem-203	75	11	on	on	ADP
fbem-203	75	12	cuda	cuda	PROPN
fbem-203	75	13	10.2	10.2	NUM
fbem-203	75	14	and	and	CCONJ
fbem-203	75	15	pytorch1.2.0	pytorch1.2.0	PROPN
fbem-203	75	16	programming	programming	NOUN
fbem-203	75	17	language	language	NOUN
fbem-203	75	18	,	,	PUNCT
fbem-203	75	19	using	use	VERB
fbem-203	75	20	five	five	NUM
fbem-203	75	21	nvidia	nvidia	PROPN
fbem-203	75	22	geforce	geforce	NOUN
fbem-203	75	23	gtx	gtx	PROPN
fbem-203	75	24	1080ti	1080ti	PROPN
fbem-203	75	25	11	11	NUM
fbem-203	75	26	g	g	NOUN
fbem-203	75	27	for	for	ADP
fbem-203	75	28	training	training	NOUN
fbem-203	75	29	,	,	PUNCT
fbem-203	75	30	and	and	CCONJ
fbem-203	75	31	the	the	DET
fbem-203	75	32	model	model	NOUN
fbem-203	75	33	is	be	AUX
fbem-203	75	34	trained	train	VERB
fbem-203	75	35	offline	offline	ADJ
fbem-203	75	36	on	on	ADP
fbem-203	75	37	the	the	DET
fbem-203	75	38	got-10k	got-10k	NOUN
fbem-203	75	39	labeled	label	VERB
fbem-203	75	40	data	datum	NOUN
fbem-203	75	41	set	set	VERB
fbem-203	75	42	;	;	PUNCT
fbem-203	75	43	the	the	DET
fbem-203	75	44	initial	initial	ADJ
fbem-203	75	45	value	value	NOUN
fbem-203	75	46	of	of	ADP
fbem-203	75	47	the	the	DET
fbem-203	75	48	learning	learning	NOUN
fbem-203	75	49	rate	rate	NOUN
fbem-203	75	50	is	be	AUX
fbem-203	75	51	0.01	0.01	NUM
fbem-203	75	52	;	;	PUNCT
fbem-203	75	53	the	the	DET
fbem-203	75	54	cross	cross	ADJ
fbem-203	75	55	-	-	ADJ
fbem-203	75	56	entropy	entropy	ADJ
fbem-203	75	57	loss	loss	NOUN
fbem-203	75	58	function	function	NOUN
fbem-203	75	59	is	be	AUX
fbem-203	75	60	used	use	VERB
fbem-203	75	61	;	;	PUNCT
fbem-203	75	62	train	train	VERB
fbem-203	75	63	for	for	ADP
fbem-203	75	64	200	200	NUM
fbem-203	75	65	epochs	epoch	NOUN
fbem-203	75	66	.	.	PUNCT
fbem-203	76	1	in	in	ADP
fbem-203	76	2	order	order	NOUN
fbem-203	76	3	to	to	PART
fbem-203	76	4	verify	verify	VERB
fbem-203	76	5	whether	whether	SCONJ
fbem-203	76	6	the	the	DET
fbem-203	76	7	effect	effect	NOUN
fbem-203	76	8	of	of	ADP
fbem-203	76	9	the	the	DET
fbem-203	76	10	algorithm	algorithm	NOUN
fbem-203	76	11	in	in	ADP
fbem-203	76	12	this	this	DET
fbem-203	76	13	paper	paper	NOUN
fbem-203	76	14	is	be	AUX
fbem-203	76	15	improved	improve	VERB
fbem-203	76	16	,	,	PUNCT
fbem-203	76	17	the	the	DET
fbem-203	76	18	otb2015	otb2015	NOUN
fbem-203	76	19	data	datum	NOUN
fbem-203	76	20	set	set	VERB
fbem-203	76	21	will	will	AUX
fbem-203	76	22	be	be	AUX
fbem-203	76	23	compared	compare	VERB
fbem-203	76	24	,	,	PUNCT
fbem-203	76	25	and	and	CCONJ
fbem-203	76	26	a	a	DET
fbem-203	76	27	representative	representative	ADJ
fbem-203	76	28	video	video	NOUN
fbem-203	76	29	sequence	sequence	NOUN
fbem-203	76	30	will	will	AUX
fbem-203	76	31	be	be	AUX
fbem-203	76	32	selected	select	VERB
fbem-203	76	33	for	for	ADP
fbem-203	76	34	quantitative	quantitative	ADJ
fbem-203	76	35	analysis	analysis	NOUN
fbem-203	76	36	.	.	PUNCT
fbem-203	77	1	4.2	4.2	NUM
fbem-203	77	2	.	.	PUNCT
fbem-203	77	3	evaluation	evaluation	NOUN
fbem-203	77	4	index	index	NOUN
fbem-203	77	5	this	this	DET
fbem-203	77	6	article	article	NOUN
fbem-203	77	7	judges	judge	VERB
fbem-203	77	8	the	the	DET
fbem-203	77	9	comprehensive	comprehensive	ADJ
fbem-203	77	10	performance	performance	NOUN
fbem-203	77	11	of	of	ADP
fbem-203	77	12	each	each	DET
fbem-203	77	13	algorithm	algorithm	NOUN
fbem-203	77	14	based	base	VERB
fbem-203	77	15	on	on	ADP
fbem-203	77	16	the	the	DET
fbem-203	77	17	accuracy	accuracy	NOUN
fbem-203	77	18	and	and	CCONJ
fbem-203	77	19	success	success	NOUN
fbem-203	77	20	rate	rate	NOUN
fbem-203	77	21	of	of	ADP
fbem-203	77	22	target	target	NOUN
fbem-203	77	23	tracking	tracking	NOUN
fbem-203	77	24	.	.	PUNCT
fbem-203	78	1	(	(	PUNCT
fbem-203	78	2	2	2	X
fbem-203	78	3	)	)	PUNCT
fbem-203	78	4	as	as	SCONJ
fbem-203	78	5	shown	show	VERB
fbem-203	78	6	in	in	ADP
fbem-203	78	7	formula	formula	NOUN
fbem-203	78	8	(	(	PUNCT
fbem-203	78	9	3	3	NUM
fbem-203	78	10	):	):	PUNCT
fbem-203	78	11	the	the	DET
fbem-203	78	12	calculation	calculation	NOUN
fbem-203	78	13	of	of	ADP
fbem-203	78	14	the	the	DET
fbem-203	78	15	target	target	NOUN
fbem-203	78	16	tracking	track	VERB
fbem-203	78	17	success	success	NOUN
fbem-203	78	18	rate	rate	NOUN
fbem-203	78	19	refers	refer	VERB
fbem-203	78	20	to	to	ADP
fbem-203	78	21	the	the	DET
fbem-203	78	22	area	area	NOUN
fbem-203	78	23	of	of	ADP
fbem-203	78	24	the	the	DET
fbem-203	78	25	overlap	overlap	NOUN
fbem-203	78	26	rate	rate	NOUN
fbem-203	78	27	(	(	PUNCT
fbem-203	78	28	iou	iou	NOUN
fbem-203	78	29	)	)	PUNCT
fbem-203	78	30	curve	curve	NOUN
fbem-203	78	31	of	of	ADP
fbem-203	78	32	the	the	DET
fbem-203	78	33	target	target	NOUN
fbem-203	78	34	's	's	PART
fbem-203	78	35	prediction	prediction	NOUN
fbem-203	78	36	block	block	NOUN
fbem-203	78	37	diagram	diagram	NOUN
fbem-203	78	38	and	and	CCONJ
fbem-203	78	39	the	the	DET
fbem-203	78	40	true	true	ADJ
fbem-203	78	41	boundary	boundary	ADJ
fbem-203	78	42	block	block	NOUN
fbem-203	78	43	diagram	diagram	NOUN
fbem-203	78	44	.	.	PUNCT
fbem-203	79	1	iou	iou	PROPN
fbem-203	79	2	|	|	ADV
fbem-203	79	3	∩	∩	ADJ
fbem-203	79	4	|	|	ADV
fbem-203	79	5	|	|	ADV
fbem-203	79	6	∪	∪	ADJ
fbem-203	80	1	|	|	NOUN
fbem-203	80	2	(	(	PUNCT
fbem-203	80	3	2	2	NUM
fbem-203	80	4	)	)	PUNCT
fbem-203	80	5	the	the	PRON
fbem-203	80	6	greater	great	ADJ
fbem-203	80	7	the	the	DET
fbem-203	80	8	value	value	NOUN
fbem-203	80	9	of	of	ADP
fbem-203	80	10	iou	iou	NOUN
fbem-203	80	11	,	,	PUNCT
fbem-203	80	12	the	the	PRON
fbem-203	80	13	higher	high	ADJ
fbem-203	80	14	the	the	DET
fbem-203	80	15	success	success	NOUN
fbem-203	80	16	rate	rate	NOUN
fbem-203	80	17	.	.	PUNCT
fbem-203	81	1	when	when	SCONJ
fbem-203	81	2	iou>0.5	iou>0.5	PROPN
fbem-203	81	3	,	,	PUNCT
fbem-203	81	4	it	it	PRON
fbem-203	81	5	can	can	AUX
fbem-203	81	6	be	be	AUX
fbem-203	81	7	considered	consider	VERB
fbem-203	81	8	that	that	SCONJ
fbem-203	81	9	the	the	DET
fbem-203	81	10	target	target	NOUN
fbem-203	81	11	has	have	AUX
fbem-203	81	12	been	be	AUX
fbem-203	81	13	successfully	successfully	ADV
fbem-203	81	14	located	locate	VERB
fbem-203	81	15	.	.	PUNCT
fbem-203	82	1	(	(	PUNCT
fbem-203	82	2	2	2	X
fbem-203	82	3	)	)	PUNCT
fbem-203	82	4	the	the	DET
fbem-203	82	5	formula	formula	NOUN
fbem-203	82	6	for	for	ADP
fbem-203	82	7	calculating	calculate	VERB
fbem-203	82	8	target	target	NOUN
fbem-203	82	9	tracking	tracking	NOUN
fbem-203	82	10	accuracy	accuracy	NOUN
fbem-203	82	11	is	be	AUX
fbem-203	82	12	shown	show	VERB
fbem-203	82	13	in	in	ADP
fbem-203	82	14	figure	figure	NOUN
fbem-203	82	15	(	(	PUNCT
fbem-203	82	16	3	3	NUM
fbem-203	82	17	)	)	PUNCT
fbem-203	82	18	below	below	ADV
fbem-203	82	19	,	,	PUNCT
fbem-203	82	20	which	which	PRON
fbem-203	82	21	means	mean	VERB
fbem-203	82	22	calculating	calculate	VERB
fbem-203	82	23	the	the	DET
fbem-203	82	24	euclidean	euclidean	ADJ
fbem-203	82	25	metric	metric	NOUN
fbem-203	82	26	between	between	ADP
fbem-203	82	27	the	the	DET
fbem-203	82	28	center	center	NOUN
fbem-203	82	29	positions	position	NOUN
fbem-203	82	30	(	(	PUNCT
fbem-203	82	31	xt	xt	X
fbem-203	82	32	,	,	PUNCT
fbem-203	82	33	yt	yt	PROPN
fbem-203	82	34	)	)	PUNCT
fbem-203	82	35	and	and	CCONJ
fbem-203	82	36	(	(	PUNCT
fbem-203	82	37	xa	xa	PROPN
fbem-203	82	38	,	,	PUNCT
fbem-203	82	39	ya	ya	PROPN
fbem-203	82	40	)	)	PUNCT
fbem-203	82	41	of	of	ADP
fbem-203	82	42	rt	rt	PROPN
fbem-203	82	43	and	and	CCONJ
fbem-203	82	44	ra	ra	PROPN
fbem-203	82	45	.	.	PUNCT
fbem-203	83	1	ε	ε	PROPN
fbem-203	83	2	x	x	PUNCT
fbem-203	84	1	x	x	PUNCT
fbem-203	84	2	y	y	PROPN
fbem-203	84	3	y	y	PROPN
fbem-203	84	4	(	(	PUNCT
fbem-203	84	5	3	3	NUM
fbem-203	84	6	)	)	PUNCT
fbem-203	84	7	4.3	4.3	NUM
fbem-203	84	8	.	.	PUNCT
fbem-203	84	9	quantitative	quantitative	ADJ
fbem-203	84	10	molecules	molecule	NOUN
fbem-203	84	11	in	in	ADP
fbem-203	84	12	order	order	NOUN
fbem-203	84	13	to	to	PART
fbem-203	84	14	comprehensively	comprehensively	ADV
fbem-203	84	15	evaluate	evaluate	VERB
fbem-203	84	16	the	the	DET
fbem-203	84	17	algorithm	algorithm	NOUN
fbem-203	84	18	proposed	propose	VERB
fbem-203	84	19	in	in	ADP
fbem-203	84	20	this	this	DET
fbem-203	84	21	paper	paper	NOUN
fbem-203	84	22	,	,	PUNCT
fbem-203	84	23	tests	test	NOUN
fbem-203	84	24	are	be	AUX
fbem-203	84	25	performed	perform	VERB
fbem-203	84	26	on	on	ADP
fbem-203	84	27	the	the	DET
fbem-203	84	28	otb	otb	PROPN
fbem-203	84	29	data	datum	NOUN
fbem-203	84	30	set	set	VERB
fbem-203	84	31	to	to	PART
fbem-203	84	32	evaluate	evaluate	VERB
fbem-203	84	33	the	the	DET
fbem-203	84	34	distance	distance	NOUN
fbem-203	84	35	and	and	CCONJ
fbem-203	84	36	overlap	overlap	NOUN
fbem-203	84	37	accuracy	accuracy	NOUN
fbem-203	84	38	of	of	ADP
fbem-203	84	39	the	the	DET
fbem-203	84	40	algorithm	algorithm	NOUN
fbem-203	84	41	in	in	ADP
fbem-203	84	42	this	this	DET
fbem-203	84	43	paper	paper	NOUN
fbem-203	84	44	.	.	PUNCT
fbem-203	85	1	in	in	ADP
fbem-203	85	2	view	view	NOUN
fbem-203	85	3	of	of	ADP
fbem-203	85	4	the	the	DET
fbem-203	85	5	various	various	ADJ
fbem-203	85	6	illumination	illumination	NOUN
fbem-203	85	7	changes	change	NOUN
fbem-203	85	8	and	and	CCONJ
fbem-203	85	9	scale	scale	NOUN
fbem-203	85	10	changes	change	NOUN
fbem-203	85	11	in	in	ADP
fbem-203	85	12	the	the	DET
fbem-203	85	13	otb	otb	PROPN
fbem-203	85	14	data	datum	NOUN
fbem-203	85	15	set	set	VERB
fbem-203	85	16	,	,	PUNCT
fbem-203	85	17	table	table	NOUN
fbem-203	85	18	1	1	NUM
fbem-203	85	19	also	also	ADV
fbem-203	85	20	shows	show	VERB
fbem-203	85	21	the	the	DET
fbem-203	85	22	accuracy	accuracy	NOUN
fbem-203	85	23	and	and	CCONJ
fbem-203	85	24	tracking	track	VERB
fbem-203	85	25	success	success	NOUN
fbem-203	85	26	rate	rate	NOUN
fbem-203	85	27	of	of	ADP
fbem-203	85	28	the	the	DET
fbem-203	85	29	algorithm	algorithm	NOUN
fbem-203	85	30	in	in	ADP
fbem-203	85	31	this	this	DET
fbem-203	85	32	paper	paper	NOUN
fbem-203	85	33	in	in	ADP
fbem-203	85	34	dealing	deal	VERB
fbem-203	85	35	with	with	ADP
fbem-203	85	36	these	these	DET
fbem-203	85	37	complex	complex	ADJ
fbem-203	85	38	tracking	tracking	NOUN
fbem-203	85	39	factors	factor	NOUN
fbem-203	85	40	.	.	PUNCT
fbem-203	86	1	table	table	NOUN
fbem-203	86	2	1	1	NUM
fbem-203	86	3	.	.	PUNCT
fbem-203	87	1	quantitative	quantitative	ADJ
fbem-203	87	2	comparrisn	comparrisn	PROPN
fbem-203	87	3	of	of	ADP
fbem-203	87	4	different	different	ADJ
fbem-203	87	5	target	target	NOUN
fbem-203	87	6	trackers	tracker	NOUN
fbem-203	87	7	on	on	ADP
fbem-203	87	8	otb2015	otb2015	NOUN
fbem-203	87	9	success	success	NOUN
fbem-203	87	10	rate	rate	NOUN
fbem-203	87	11	precision	precision	NOUN
fbem-203	87	12	sa	sa	PROPN
fbem-203	87	13	-	-	PUNCT
fbem-203	87	14	resdpp	resdpp	NOUN
fbem-203	87	15	-	-	PUNCT
fbem-203	87	16	siamfc	siamfc	NOUN
fbem-203	87	17	0.512	0.512	NUM
fbem-203	87	18	0.522	0.522	NUM
fbem-203	87	19	resdpp	resdpp	NOUN
fbem-203	87	20	-	-	PUNCT
fbem-203	87	21	siamfc	siamfc	NOUN
fbem-203	87	22	0.484	0.484	NUM
fbem-203	87	23	0.496	0.496	NUM
fbem-203	87	24	dpp	dpp	NOUN
fbem-203	87	25	-	-	PUNCT
fbem-203	87	26	siamfc	siamfc	NOUN
fbem-203	87	27	0.479	0.479	NUM
fbem-203	87	28	0.488	0.488	NUM
fbem-203	87	29	siamfc	siamfc	NOUN
fbem-203	87	30	0.456	0.456	NUM
fbem-203	87	31	0.478	0.478	NUM
fbem-203	87	32	5	5	NUM
fbem-203	87	33	.	.	PUNCT
fbem-203	88	1	summary	summary	NOUN
fbem-203	88	2	and	and	CCONJ
fbem-203	88	3	outlook	outlook	NOUN
fbem-203	88	4	in	in	ADP
fbem-203	88	5	this	this	DET
fbem-203	88	6	paper	paper	NOUN
fbem-203	88	7	,	,	PUNCT
fbem-203	88	8	the	the	DET
fbem-203	88	9	siamfc	siamfc	NOUN
fbem-203	88	10	algorithm	algorithm	NOUN
fbem-203	88	11	is	be	AUX
fbem-203	88	12	improved	improve	VERB
fbem-203	88	13	and	and	CCONJ
fbem-203	88	14	optimized	optimize	VERB
fbem-203	88	15	.	.	PUNCT
fbem-203	89	1	first	first	ADV
fbem-203	89	2	,	,	PUNCT
fbem-203	89	3	the	the	DET
fbem-203	89	4	detailed	detailed	ADJ
fbem-203	89	5	feature	feature	NOUN
fbem-203	89	6	extraction	extraction	NOUN
fbem-203	89	7	is	be	AUX
fbem-203	89	8	introduced	introduce	VERB
fbem-203	89	9	into	into	ADP
fbem-203	89	10	the	the	DET
fbem-203	89	11	vgg	vgg	NOUN
fbem-203	89	12	network	network	NOUN
fbem-203	89	13	,	,	PUNCT
fbem-203	89	14	and	and	CCONJ
fbem-203	89	15	the	the	DET
fbem-203	89	16	residual	residual	ADJ
fbem-203	89	17	network	network	NOUN
fbem-203	89	18	is	be	AUX
fbem-203	89	19	used	use	VERB
fbem-203	89	20	,	,	PUNCT
fbem-203	89	21	so	so	SCONJ
fbem-203	89	22	that	that	SCONJ
fbem-203	89	23	the	the	DET
fbem-203	89	24	detailed	detailed	ADJ
fbem-203	89	25	feature	feature	NOUN
fbem-203	89	26	retention	retention	NOUN
fbem-203	89	27	module	module	NOUN
fbem-203	89	28	is	be	AUX
fbem-203	89	29	embedded	embed	VERB
fbem-203	89	30	in	in	ADP
fbem-203	89	31	each	each	DET
fbem-203	89	32	layer	layer	NOUN
fbem-203	89	33	,	,	PUNCT
fbem-203	89	34	and	and	CCONJ
fbem-203	89	35	the	the	DET
fbem-203	89	36	change	change	NOUN
fbem-203	89	37	of	of	ADP
fbem-203	89	38	the	the	DET
fbem-203	89	39	target	target	NOUN
fbem-203	89	40	feature	feature	NOUN
fbem-203	89	41	is	be	AUX
fbem-203	89	42	enlarged	enlarge	VERB
fbem-203	89	43	and	and	CCONJ
fbem-203	89	44	retained	retain	VERB
fbem-203	89	45	.	.	PUNCT
fbem-203	90	1	the	the	DET
fbem-203	90	2	important	important	ADJ
fbem-203	90	3	structural	structural	ADJ
fbem-203	90	4	details	detail	NOUN
fbem-203	90	5	of	of	ADP
fbem-203	90	6	the	the	DET
fbem-203	90	7	original	original	ADJ
fbem-203	90	8	target	target	NOUN
fbem-203	90	9	feature	feature	NOUN
fbem-203	90	10	play	play	VERB
fbem-203	90	11	a	a	DET
fbem-203	90	12	certain	certain	ADJ
fbem-203	90	13	role	role	NOUN
fbem-203	90	14	in	in	ADP
fbem-203	90	15	the	the	DET
fbem-203	90	16	search	search	NOUN
fbem-203	90	17	and	and	CCONJ
fbem-203	90	18	discrimination	discrimination	NOUN
fbem-203	90	19	of	of	ADP
fbem-203	90	20	the	the	DET
fbem-203	90	21	details	detail	NOUN
fbem-203	90	22	in	in	ADP
fbem-203	90	23	the	the	DET
fbem-203	90	24	target	target	NOUN
fbem-203	90	25	tracking	tracking	NOUN
fbem-203	90	26	,	,	PUNCT
fbem-203	90	27	and	and	CCONJ
fbem-203	90	28	solve	solve	VERB
fbem-203	90	29	the	the	DET
fbem-203	90	30	problem	problem	NOUN
fbem-203	90	31	of	of	ADP
fbem-203	90	32	the	the	DET
fbem-203	90	33	original	original	ADJ
fbem-203	90	34	tracking	tracking	NOUN
fbem-203	90	35	network	network	NOUN
fbem-203	90	36	losing	lose	VERB
fbem-203	90	37	the	the	DET
fbem-203	90	38	target	target	NOUN
fbem-203	90	39	details	detail	NOUN
fbem-203	90	40	when	when	SCONJ
fbem-203	90	41	the	the	DET
fbem-203	90	42	parameter	parameter	NOUN
fbem-203	90	43	scale	scale	NOUN
fbem-203	90	44	is	be	AUX
fbem-203	90	45	reduced	reduce	VERB
fbem-203	90	46	.	.	PUNCT
fbem-203	91	1	finally	finally	ADV
fbem-203	91	2	,	,	PUNCT
fbem-203	91	3	the	the	DET
fbem-203	91	4	spatial	spatial	ADJ
fbem-203	91	5	attention	attention	NOUN
fbem-203	91	6	mechanism	mechanism	NOUN
fbem-203	91	7	is	be	AUX
fbem-203	91	8	used	use	VERB
fbem-203	91	9	.	.	PUNCT
fbem-203	92	1	to	to	PART
fbem-203	92	2	improve	improve	VERB
fbem-203	92	3	the	the	DET
fbem-203	92	4	network	network	NOUN
fbem-203	92	5	model	model	NOUN
fbem-203	92	6	's	's	PART
fbem-203	92	7	attention	attention	NOUN
fbem-203	92	8	to	to	ADP
fbem-203	92	9	effective	effective	ADJ
fbem-203	92	10	features	feature	NOUN
fbem-203	92	11	during	during	ADP
fbem-203	92	12	feature	feature	NOUN
fbem-203	92	13	extraction	extraction	NOUN
fbem-203	92	14	.	.	PUNCT
fbem-203	93	1	although	although	SCONJ
fbem-203	93	2	the	the	DET
fbem-203	93	3	accuracy	accuracy	NOUN
fbem-203	93	4	and	and	CCONJ
fbem-203	93	5	success	success	NOUN
fbem-203	93	6	rate	rate	NOUN
fbem-203	93	7	in	in	ADP
fbem-203	93	8	target	target	NOUN
fbem-203	93	9	tracking	tracking	NOUN
fbem-203	93	10	have	have	AUX
fbem-203	93	11	been	be	AUX
fbem-203	93	12	improved	improve	VERB
fbem-203	93	13	,	,	PUNCT
fbem-203	93	14	there	there	PRON
fbem-203	93	15	are	be	VERB
fbem-203	93	16	still	still	ADV
fbem-203	93	17	some	some	DET
fbem-203	93	18	shortcomings	shortcoming	NOUN
fbem-203	93	19	in	in	ADP
fbem-203	93	20	real	real	ADJ
fbem-203	93	21	-	-	PUNCT
fbem-203	93	22	time	time	NOUN
fbem-203	93	23	performance	performance	NOUN
fbem-203	93	24	,	,	PUNCT
fbem-203	93	25	and	and	CCONJ
fbem-203	93	26	the	the	DET
fbem-203	93	27	real	real	ADJ
fbem-203	93	28	-	-	PUNCT
fbem-203	93	29	time	time	NOUN
fbem-203	93	30	problem	problem	NOUN
fbem-203	93	31	in	in	ADP
fbem-203	93	32	target	target	NOUN
fbem-203	93	33	tracking	tracking	NOUN
fbem-203	93	34	needs	need	NOUN
fbem-203	93	35	to	to	PART
fbem-203	93	36	be	be	AUX
fbem-203	93	37	solved	solve	VERB
fbem-203	93	38	.	.	PUNCT
fbem-203	94	1	references	reference	NOUN
fbem-203	94	2	[	[	X
fbem-203	94	3	1	1	X
fbem-203	94	4	]	]	PUNCT
fbem-203	94	5	peng	peng	PROPN
fbem-203	94	6	j	j	PROPN
fbem-203	94	7	y	y	PROPN
fbem-203	94	8	,	,	PUNCT
fbem-203	94	9	chen	chen	PROPN
fbem-203	94	10	x	x	SYM
fbem-203	94	11	b.	b.	PROPN
fbem-203	94	12	novel	novel	NOUN
fbem-203	94	13	models	model	NOUN
fbem-203	94	14	for	for	ADP
fbem-203	94	15	one	one	NUM
fbem-203	94	16	-	-	PUNCT
fbem-203	94	17	sided	sided	ADJ
fbem-203	94	18	hysteresis	hysteresis	NOUN
fbem-203	94	19	of	of	ADP
fbem-203	94	20	piezoelectric	piezoelectric	ADJ
fbem-203	94	21	actuator	actuator	NOUN
fbem-203	95	1	[	[	X
fbem-203	95	2	j	j	X
fbem-203	95	3	]	]	X
fbem-203	95	4	.	.	PUNCT
fbem-203	96	1	mechatronics	mechatronic	NOUN
fbem-203	96	2	,	,	PUNCT
fbem-203	96	3	2012	2012	NUM
fbem-203	96	4	,	,	PUNCT
fbem-203	96	5	22(6):757765	22(6):757765	NUM
fbem-203	96	6	.	.	PUNCT
fbem-203	97	1	[	[	X
fbem-203	97	2	2	2	NUM
fbem-203	97	3	]	]	X
fbem-203	97	4	yang	yang	PROPN
fbem-203	97	5	kang	kang	PROPN
fbem-203	97	6	,	,	PUNCT
fbem-203	97	7	song	song	PROPN
fbem-203	97	8	huihui	huihui	PROPN
fbem-203	97	9	,	,	PUNCT
fbem-203	97	10	zhang	zhang	PROPN
fbem-203	97	11	kaihua	kaihua	PROPN
fbem-203	97	12	.	.	PUNCT
fbem-203	98	1	real	real	ADJ
fbem-203	98	2	-	-	PUNCT
fbem-203	98	3	time	time	NOUN
fbem-203	98	4	visual	visual	ADJ
fbem-203	98	5	tracking	tracking	NOUN
fbem-203	98	6	based	base	VERB
fbem-203	98	7	on	on	ADP
fbem-203	98	8	dual	dual	ADJ
fbem-203	98	9	attention	attention	NOUN
fbem-203	98	10	siamese	siamese	ADJ
fbem-203	98	11	network	network	NOUN
fbem-203	99	1	[	[	X
fbem-203	99	2	j	j	X
fbem-203	99	3	]	]	X
fbem-203	99	4	.	.	PUNCT
fbem-203	100	1	journal	journal	PROPN
fbem-203	100	2	of	of	ADP
fbem-203	100	3	computer	computer	NOUN
fbem-203	100	4	applications	application	NOUN
fbem-203	100	5	,	,	PUNCT
fbem-203	100	6	2019	2019	NUM
fbem-203	100	7	,	,	PUNCT
fbem-203	100	8	39(6):1652	39(6):1652	NUM
fbem-203	100	9	-	-	SYM
fbem-203	100	10	1656	1656	NUM
fbem-203	100	11	.	.	PUNCT
fbem-203	101	1	(	(	PUNCT
fbem-203	101	2	in	in	ADP
fbem-203	101	3	chinese	chinese	PROPN
fbem-203	101	4	)	)	PUNCT
fbem-203	102	1	[	[	X
fbem-203	102	2	3	3	X
fbem-203	102	3	]	]	X
fbem-203	102	4	kuai	kuai	X
fbem-203	102	5	y	y	PROPN
fbem-203	102	6	,	,	PUNCT
fbem-203	102	7	weng	weng	PROPN
fbem-203	102	8	,	,	PUNCT
fbem-203	102	9	li	li	PROPN
fbem-203	102	10	d.hyper	d.hyper	VERB
fbem-203	102	11	-	-	ADJ
fbem-203	102	12	siamese	siamese	ADJ
fbem-203	102	13	network	network	NOUN
fbem-203	102	14	forrobust	forrobust	NOUN
fbem-203	102	15	visual	visual	ADJ
fbem-203	102	16	tracking[j	tracking[j	NOUN
fbem-203	102	17	]	]	PUNCT
fbem-203	102	18	.	.	PUNCT
fbem-203	103	1	signal	signal	NOUN
fbem-203	103	2	,	,	PUNCT
fbem-203	103	3	image	image	NOUN
fbem-203	103	4	and	and	CCONJ
fbem-203	103	5	video	video	NOUN
fbem-203	103	6	processing	processing	NOUN
fbem-203	103	7	,	,	PUNCT
fbem-203	103	8	2019	2019	NUM
fbem-203	103	9	,	,	PUNCT
fbem-203	103	10	1	1	NUM
fbem-203	103	11	3(1):35	3(1):35	NUM
fbem-203	103	12	-	-	SYM
fbem-203	103	13	42	42	NUM
fbem-203	103	14	.	.	PUNCT
fbem-203	104	1	[	[	X
fbem-203	104	2	4	4	NUM
fbem-203	104	3	]	]	PUNCT
fbem-203	104	4	henriques	henrique	NOUN
fbem-203	105	1	j	j	PROPN
fbem-203	105	2	f	f	X
fbem-203	105	3	,	,	PUNCT
fbem-203	105	4	caseiro	caseiro	PROPN
fbem-203	105	5	r	r	PROPN
fbem-203	105	6	,	,	PUNCT
fbem-203	105	7	martins	martin	VERB
fbem-203	105	8	p	p	NOUN
fbem-203	105	9	,	,	PUNCT
fbem-203	105	10	et	et	PROPN
fbem-203	105	11	al	al	PROPN
fbem-203	105	12	.	.	PROPN
fbem-203	105	13	high	high	ADJ
fbem-203	105	14	-	-	PUNCT
fbem-203	105	15	speed	speed	NOUN
fbem-203	105	16	tracking	tracking	NOUN
fbem-203	105	17	with	with	ADP
fbem-203	105	18	kernelized	kernelize	VERB
fbem-203	105	19	correlation	correlation	NOUN
fbem-203	105	20	filters[j	filters[j	PROPN
fbem-203	105	21	]	]	PUNCT
fbem-203	105	22	.	.	PUNCT
fbem-203	106	1	ieee	ieee	NOUN
fbem-203	106	2	transactions	transaction	NOUN
fbem-203	106	3	on	on	ADP
fbem-203	106	4	pattern	pattern	NOUN
fbem-203	106	5	analysis	analysis	NOUN
fbem-203	106	6	and	and	CCONJ
fbem-203	106	7	machineintelli	machineintelli	NOUN
fbem-203	106	8	-gence	-gence	PROPN
fbem-203	106	9	,	,	PUNCT
fbem-203	106	10	2015	2015	NUM
fbem-203	106	11	,	,	PUNCT
fbem-203	106	12	37(3	37(3	NUM
fbem-203	106	13	):	):	PUNCT
fbem-203	106	14	583596	583596	NUM
fbem-203	106	15	.	.	PUNCT
fbem-203	107	1	[	[	X
fbem-203	107	2	5	5	NUM
fbem-203	107	3	]	]	PUNCT
fbem-203	107	4	xu	xu	PROPN
fbem-203	107	5	t	t	PROPN
fbem-203	107	6	,	,	PUNCT
fbem-203	107	7	feng	feng	PROPN
fbem-203	107	8	z	z	PROPN
fbem-203	107	9	h	h	PROPN
fbem-203	107	10	,	,	PUNCT
fbem-203	107	11	wu	wu	PROPN
fbem-203	107	12	x	x	PROPN
fbem-203	107	13	,	,	PUNCT
fbem-203	107	14	etal	etal	NOUN
fbem-203	107	15	.	.	PUNCT
fbem-203	108	1	learning	learn	VERB
fbem-203	108	2	adaptive	adaptive	ADJ
fbem-203	108	3	discriminative	discriminative	NOUN
fbem-203	108	4	correlation	correlation	NOUN
fbem-203	108	5	filters	filter	NOUN
fbem-203	108	6	via	via	ADP
fbem-203	108	7	temporal	temporal	ADJ
fbem-203	108	8	consistency	consistency	NOUN
fbem-203	108	9	preserving	preserve	VERB
fbem-203	108	10	spatial	spatial	ADJ
fbem-203	108	11	feature	feature	NOUN
fbem-203	108	12	selection	selection	NOUN
fbem-203	108	13	for	for	ADP
fbem-203	108	14	robust	robust	ADJ
fbem-203	108	15	visual	visual	ADJ
fbem-203	108	16	object	object	NOUN
fbem-203	108	17	tracking	track	VERB
fbem-203	109	1	[	[	X
fbem-203	109	2	j	j	X
fbem-203	109	3	]	]	X
fbem-203	109	4	.	.	PUNCT
fbem-203	110	1	ieee	ieee	NOUN
fbem-203	110	2	transactions	transaction	NOUN
fbem-203	110	3	on	on	ADP
fbem-203	110	4	image	image	NOUN
fbem-203	110	5	processing	processing	NOUN
fbem-203	110	6	,	,	PUNCT
fbem-203	110	7	2019	2019	NUM
fbem-203	110	8	,	,	PUNCT
fbem-203	110	9	28(11):5596	28(11):5596	NUM
fbem-203	110	10	-	-	SYM
fbem-203	110	11	5609	5609	NUM
fbem-203	110	12	.	.	PUNCT
fbem-203	111	1	[	[	X
fbem-203	111	2	6	6	NUM
fbem-203	111	3	]	]	X
fbem-203	111	4	abbas	abbas	PROPN
fbem-203	111	5	manuel	manuel	PROPN
fbem-203	111	6	,	,	PUNCT
fbem-203	111	7	somme	somme	PROPN
fbem-203	111	8	dominique	dominique	PROPN
fbem-203	111	9	,	,	PUNCT
fbem-203	111	10	le	le	X
fbem-203	111	11	bouquin	bouquin	PROPN
fbem-203	111	12	jeannès	jeannès	PROPN
fbem-203	111	13	régine	régine	PROPN
fbem-203	111	14	.	.	PUNCT
fbem-203	112	1	d	d	X
fbem-203	112	2	-	-	PUNCT
fbem-203	112	3	sorm	sorm	NOUN
fbem-203	112	4	:	:	PUNCT
fbem-203	112	5	a	a	DET
fbem-203	112	6	digital	digital	ADJ
fbem-203	112	7	solution	solution	NOUN
fbem-203	112	8	for	for	ADP
fbem-203	112	9	remote	remote	ADJ
fbem-203	112	10	monitoring	monitoring	NOUN
fbem-203	112	11	based	base	VERB
fbem-203	112	12	on	on	ADP
fbem-203	112	13	the	the	DET
fbem-203	112	14	attitude	attitude	NOUN
fbem-203	112	15	of	of	ADP
fbem-203	112	16	wearable	wearable	ADJ
fbem-203	112	17	devices[j	devices[j	PROPN
fbem-203	112	18	]	]	PUNCT
fbem-203	112	19	.	.	PUNCT
fbem-203	113	1	computer	computer	NOUN
fbem-203	113	2	methods	method	NOUN
fbem-203	113	3	and	and	CCONJ
fbem-203	113	4	programs	program	NOUN
fbem-203	113	5	in	in	ADP
fbem-203	113	6	biomedicine,2021,208	biomedicine,2021,208	NOUN
fbem-203	113	7	:	:	PUNCT
fbem-203	114	1	[	[	X
fbem-203	114	2	7	7	X
fbem-203	114	3	]	]	X
fbem-203	114	4	[	[	X
fbem-203	114	5	7	7	NUM
fbem-203	114	6	]	]	X
fbem-203	114	7	bai	bai	NOUN
fbem-203	114	8	xingzhen	xingzhen	PROPN
fbem-203	114	9	,	,	PUNCT
fbem-203	114	10	et	et	PROPN
fbem-203	114	11	al	al	PROPN
fbem-203	114	12	.	.	PUNCT
fbem-203	114	13	"target	"target	PROPN
fbem-203	114	14	tracking	track	VERB
fbem-203	114	15	for	for	ADP
fbem-203	114	16	wireless	wireless	ADJ
fbem-203	114	17	localization	localization	NOUN
fbem-203	114	18	systems	system	NOUN
fbem-203	114	19	using	use	VERB
fbem-203	114	20	set	set	NOUN
fbem-203	114	21	-	-	PUNCT
fbem-203	114	22	membership	membership	NOUN
fbem-203	114	23	filtering	filtering	NOUN
fbem-203	114	24	:	:	PUNCT
fbem-203	114	25	a	a	DET
fbem-203	114	26	component	component	NOUN
fbem-203	114	27	-	-	PUNCT
fbem-203	114	28	based	base	VERB
fbem-203	114	29	event	event	NOUN
fbem-203	114	30	-	-	PUNCT
fbem-203	114	31	triggered	trigger	VERB
fbem-203	114	32	mechanism	mechanism	NOUN
fbem-203	114	33	.	.	PUNCT
fbem-203	114	34	"	"	PUNCT
fbem-203	115	1	automatica	automatica	PROPN
fbem-203	115	2	132.(2021	132.(2021	NUM
fbem-203	115	3	):	):	PUNCT
fbem-203	115	4	doi:10.1016	doi:10.1016	PROPN
fbem-203	115	5	/	/	SYM
fbem-203	115	6	j.automatica.2021.109795	j.automatica.2021.109795	PROPN
fbem-203	115	7	.	.	PUNCT
fbem-203	116	1	[	[	X
fbem-203	116	2	8	8	NUM
fbem-203	116	3	]	]	PUNCT
fbem-203	116	4	bednarz	bednarz	NOUN
fbem-203	116	5	bryan	bryan	PROPN
fbem-203	116	6	p.	p.	PROPN
fbem-203	116	7	,jupitz	,jupitz	PUNCT
fbem-203	116	8	sydney	sydney	PROPN
fbem-203	116	9	,	,	PUNCT
fbem-203	116	10	lee	lee	PROPN
fbem-203	116	11	warren	warren	PROPN
fbem-203	116	12	,	,	PUNCT
fbem-203	116	13	mills	mills	PROPN
fbem-203	116	14	david	david	PROPN
fbem-203	116	15	,	,	PUNCT
fbem-203	116	16	chan	chan	PROPN
fbem-203	116	17	heather	heather	PROPN
fbem-203	116	18	,	,	PUNCT
fbem-203	116	19	fiorillo	fiorillo	PROPN
fbem-203	116	20	timothy	timothy	PROPN
fbem-203	116	21	,	,	PUNCT
fbem-203	116	22	sabitini	sabitini	PROPN
fbem-203	116	23	james	james	PROPN
fbem-203	116	24	,	,	PUNCT
fbem-203	116	25	shoudy	shoudy	PROPN
fbem-203	116	26	david	david	PROPN
fbem-203	116	27	,	,	PUNCT
fbem-203	116	28	patel	patel	PROPN
fbem-203	116	29	aqsa	aqsa	PROPN
fbem-203	116	30	,	,	PUNCT
fbem-203	116	31	mitra	mitra	PROPN
fbem-203	116	32	jhimli	jhimli	PROPN
fbem-203	116	33	,	,	PUNCT
fbem-203	116	34	sarcar	sarcar	PROPN
fbem-203	116	35	shourya	shourya	NOUN
fbem-203	116	36	,	,	PUNCT
fbem-203	116	37	wang	wang	PROPN
fbem-203	116	38	bo	bo	PROPN
fbem-203	116	39	,	,	PUNCT
fbem-203	116	40	shepard	shepard	NOUN
fbem-203	116	41	andrew	andrew	PROPN
fbem-203	116	42	,	,	PUNCT
fbem-203	116	43	matrosic	matrosic	NOUN
fbem-203	116	44	charles	charle	NOUN
fbem-203	116	45	,	,	PUNCT
fbem-203	116	46	holmes	holmes	PROPN
fbem-203	116	47	james	james	PROPN
fbem-203	116	48	,	,	PUNCT
fbem-203	116	49	culberson	culberson	NOUN
fbem-203	116	50	wesley	wesley	NOUN
fbem-203	116	51	,	,	PUNCT
fbem-203	116	52	bassetti	bassetti	NOUN
fbem-203	116	53	michael	michael	PROPN
fbem-203	116	54	,	,	PUNCT
fbem-203	116	55	hill	hill	PROPN
fbem-203	116	56	patrick	patrick	PROPN
fbem-203	116	57	,	,	PUNCT
fbem-203	116	58	mcmillan	mcmillan	PROPN
fbem-203	116	59	81	81	NUM
fbem-203	116	60	alan	alan	PROPN
fbem-203	116	61	,	,	PUNCT
fbem-203	116	62	zagzebski	zagzebski	PROPN
fbem-203	116	63	james	james	PROPN
fbem-203	116	64	,	,	PUNCT
fbem-203	116	65	smith	smith	PROPN
fbem-203	116	66	l.	l.	PROPN
fbem-203	116	67	scott	scott	PROPN
fbem-203	116	68	,	,	PUNCT
fbem-203	116	69	foo	foo	PROPN
fbem-203	116	70	thomas	thomas	PROPN
fbem-203	116	71	k	k	PROPN
fbem-203	116	72	..	..	PUNCT
fbem-203	116	73	first	first	ADJ
fbem-203	116	74	-	-	PUNCT
fbem-203	116	75	inhuman	inhuman	ADJ
fbem-203	116	76	imaging	imaging	NOUN
fbem-203	116	77	using	use	VERB
fbem-203	116	78	a	a	DET
fbem-203	116	79	mr	mr	PROPN
fbem-203	116	80	-	-	PUNCT
fbem-203	116	81	compatible	compatible	ADJ
fbem-203	116	82	e4d	e4d	NOUN
fbem-203	116	83	ultrasound	ultrasound	NOUN
fbem-203	116	84	probe	probe	NOUN
fbem-203	116	85	for	for	ADP
fbem-203	116	86	motion	motion	NOUN
fbem-203	116	87	management	management	NOUN
fbem-203	116	88	of	of	ADP
fbem-203	116	89	radiotherapy[j	radiotherapy[j	NOUN
fbem-203	116	90	]	]	PUNCT
fbem-203	116	91	.	.	PUNCT
fbem-203	117	1	physica	physica	PROPN
fbem-203	117	2	medica	medica	PROPN
fbem-203	117	3	,	,	PUNCT
fbem-203	117	4	2021	2021	NUM
fbem-203	117	5	,	,	PUNCT
fbem-203	117	6	88	88	NUM
fbem-203	117	7	:	:	PUNCT
fbem-203	118	1	[	[	X
fbem-203	118	2	9	9	NUM
fbem-203	118	3	]	]	PUNCT
fbem-203	118	4	elgamoudi	elgamoudi	PROPN
fbem-203	118	5	abulasad	abulasad	ADV
fbem-203	118	6	,	,	PUNCT
fbem-203	118	7	benzerrouk	benzerrouk	NOUN
fbem-203	118	8	hamza	hamza	PROPN
fbem-203	118	9	,	,	PUNCT
fbem-203	118	10	elango	elango	PROPN
fbem-203	118	11	g.	g.	PROPN
fbem-203	118	12	arul	arul	PROPN
fbem-203	118	13	,	,	PUNCT
fbem-203	118	14	landry	landry	PROPN
fbem-203	118	15	rené	rené	PROPN
fbem-203	118	16	.	.	PUNCT
fbem-203	119	1	a	a	DET
fbem-203	119	2	survey	survey	NOUN
fbem-203	119	3	for	for	ADP
fbem-203	119	4	recent	recent	ADJ
fbem-203	119	5	techniques	technique	NOUN
fbem-203	119	6	and	and	CCONJ
fbem-203	119	7	algorithms	algorithm	NOUN
fbem-203	119	8	of	of	ADP
fbem-203	119	9	geolocation	geolocation	NOUN
fbem-203	119	10	and	and	CCONJ
fbem-203	119	11	target	target	NOUN
fbem-203	119	12	tracking	tracking	NOUN
fbem-203	119	13	in	in	ADP
fbem-203	119	14	wireless	wireless	NOUN
fbem-203	119	15	and	and	CCONJ
fbem-203	119	16	satellite	satellite	NOUN
fbem-203	119	17	systems[j	systems[j	PROPN
fbem-203	119	18	]	]	PUNCT
fbem-203	119	19	.	.	PUNCT
fbem-203	120	1	applied	apply	VERB
fbem-203	120	2	sciences,2021,11(13	sciences,2021,11(13	PROPN
fbem-203	120	3	):	):	PUNCT
fbem-203	120	4	[	[	X
fbem-203	120	5	10	10	NUM
fbem-203	120	6	]	]	X
fbem-203	120	7	sun	sun	PROPN
fbem-203	120	8	peng	peng	PROPN
fbem-203	120	9	,	,	PUNCT
fbem-203	120	10	zhu	zhu	PROPN
fbem-203	120	11	bing	bing	NOUN
fbem-203	120	12	,	,	PUNCT
fbem-203	120	13	zuo	zuo	PROPN
fbem-203	120	14	zongyu	zongyu	PROPN
fbem-203	120	15	,	,	PUNCT
fbem-203	120	16	basin	basin	NOUN
fbem-203	120	17	michael	michael	PROPN
fbem-203	120	18	v	v	PROPN
fbem-203	120	19	..	..	PUNCT
fbem-203	120	20	visionbased	visionbase	VERB
fbem-203	120	21	finite	finite	ADJ
fbem-203	120	22	-	-	PUNCT
fbem-203	120	23	time	time	NOUN
fbem-203	120	24	uncooperative	uncooperative	ADJ
fbem-203	120	25	target	target	NOUN
fbem-203	120	26	tracking	tracking	NOUN
fbem-203	120	27	for	for	ADP
fbem-203	120	28	uav	uav	PROPN
fbem-203	120	29	subject	subject	NOUN
fbem-203	120	30	to	to	ADP
fbem-203	120	31	actuator	actuator	NOUN
fbem-203	120	32	saturation[j	saturation[j	PROPN
fbem-203	120	33	]	]	PUNCT
fbem-203	120	34	.	.	PUNCT
fbem-203	121	1	automatica,2021,130	automatica,2021,130	PROPN
fbem-203	121	2	.	.	PUNCT