Hidden Two-Stream Convolutional Networks for Action Recognition

  title={Hidden Two-Stream Convolutional Networks for Action Recognition},
  author={Yi Zhu and Zhen-Zhong Lan and Shawn D. Newsam and Alexander G. Hauptmann},
Analyzing videos of human actions involves understanding the temporal relationships among video frames. State-of-the-art action recognition approaches rely on traditional optical flow estimation methods to pre-compute motion information for CNNs. Such a two-stage approach is computationally expensive, storage demanding, and not endto-end trainable. In this paper, we present a novel CNN architecture that implicitly captures motion information between adjacent frames. We name our approach hidden… CONTINUE READING
Highly Cited
This paper has 30 citations. REVIEW CITATIONS
Related Discussions
This paper has been referenced on Twitter 23 times. VIEW TWEETS


Publications citing this paper.
Showing 1-10 of 22 extracted citations

Affective State Recognition Based on Eye Gaze Analysis Using Two–stream Convolutional Networks

2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP) • 2018
View 1 Excerpt

Fully-Coupled Two-Stream Spatiotemporal Networks for Extremely Low Resolution Action Recognition

2018 IEEE Winter Conference on Applications of Computer Vision (WACV) • 2018
View 1 Excerpt

Similar Papers

Loading similar papers…