Convolutional neural random fields for action recognition

作者：

Highlights：

• We develop a novel spatio-temporal CNN architecture for feature learning from video frames.

• CRF is coupled with CNN to achieve feature learning and structured prediction simultaneously.

• We explore different combinations of feature functions for sequence labeling.

• We validate our framework on segmented and unsegmented action datasets, respectively.

摘要

Highlights•We develop a novel spatio-temporal CNN architecture for feature learning from video frames.•CRF is coupled with CNN to achieve feature learning and structured prediction simultaneously.•We explore different combinations of feature functions for sequence labeling.•We validate our framework on segmented and unsegmented action datasets, respectively.

论文关键词：Action recognition,Sequence labeling,Conditional random fields,Spatio-temporal convolution

论文评审过程：Received 2 August 2015, Revised 13 March 2016, Accepted 14 March 2016, Available online 19 March 2016, Version of Record 23 August 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.03.019