id	author	title	date	pages	extension	mime	words	sentence	flesch	summary	cache	txt
assc-56657	Zhang, Mingzhu	A Review of Spatio-Temporal Action Detection Models in Computer Vision Based on Deep Learning	2025	12	.pdf	application/pdf	4481	197	38	With the breakthroughs in deep learning, methods based on architectures like Convolutional Neural Networks (CNNs) and Transformers have enabled automatic learning of action features, significantly improving detection accuracy and generalization ability. As a key task in video understanding, spatiotemporal action detection aims to simultaneously recognize action categories, localize spatial positions, and detect temporal intervals from continuous video sequences, integrating the technical capabilities of object detection, action recognition, and temporal analysis.	cache/assc-56657.pdf	txt/assc-56657.txt