Hidden Two-Stream Convolutional Networks for Action Recognition

TL;DR

This paper introduces a novel end-to-end CNN architecture that implicitly captures motion for action recognition directly from raw video frames, significantly improving speed and accuracy over traditional optical flow-based methods.

Contribution

The proposed hidden two-stream CNN architecture eliminates the need for explicit optical flow computation, enabling faster and more accurate action recognition from raw videos.

Findings

10x faster than two-stage optical flow methods

Significantly outperforms previous real-time approaches

Effective on multiple challenging datasets

Abstract

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 end-to-end trainable. In this paper, we present a novel CNN architecture that implicitly captures motion information between adjacent frames. We name our approach hidden two-stream CNNs because it only takes raw video frames as input and directly predicts action classes without explicitly computing optical flow. Our end-to-end approach is 10x faster than its two-stage baseline. Experimental results on four challenging action recognition datasets: UCF101, HMDB51, THUMOS14 and ActivityNet v1.2 show that our approach significantly outperforms the previous best real-time approaches.