Cascaded Temporal Spatial Features for Video Action Recognition

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 先进时空数据分析与学习

	Cascaded Temporal Spatial Features for Video Action Recognition
	Tingzhao Yu1,2 ; Huxiang Gu1 ; Lingfeng Wang1 ; Shiming Xiang and 1; Chunhong Pan1
	2017
会议名称	IEEE International Conference on Image Processing
会议日期	2017-9-17
会议地点	Beijing, CHINA
摘要	Extracting spatial-temporal descriptors is a challenging task for video-based human action recognition. We decouple the 3D volume of video frames directly into a cascaded temporal spatial domain via a new convolutional architecture. The motivation behind this design is to achieve deep nonlinear feature representations with reduced network parameters. First, a 1D temporal network with shared parameters is first constructed to map the video sequences along the time axis into feature maps in temporal domain. These feature maps are then organized into channels like those of RGB image (named as Motion Image here for abbreviation), which is desired to preserve both temporal and spatial information. Second, the Motion Image is regarded as the input of the latter cascaded 2D spatial network. With the combination of the 1D temporal network and the 2D spatial network together, the size of whole network parameters is largely reduced. Benefiting from the Motion Image, our network is an end-to-end system for the task of action recognition, which can be trained with the classical algorithm of back propagation. Quantities of comparative experiments on two benchmark datasets demonstrate the effectiveness of our new architecture.
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/20351
专题	多模态人工智能系统全国重点实验室_先进时空数据分析与学习
作者单位	1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.School of Computer and Control Engineering, University of Chinese Academy of Sciences
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Tingzhao Yu,Huxiang Gu,Lingfeng Wang,et al. Cascaded Temporal Spatial Features for Video Action Recognition[C],2017.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
19_0001552_ICIP_Yuti（4989KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载