End-to-End Learning of Motion Representation for Video Understanding

TL;DR

This paper introduces TVNet, an end-to-end trainable neural network that learns optical-flow-like features directly from data, improving video understanding tasks by integrating feature extraction into the learning process.

Contribution

The paper proposes TVNet, a neural network that unrolls the TV-L1 optical flow optimization as layers, enabling end-to-end training and task-specific feature learning for video analysis.

Findings

TVNet outperforms existing methods on action recognition benchmarks.

It achieves comparable accuracy with faster feature extraction.

End-to-end training improves task-specific feature learning.

Abstract

Despite the recent success of end-to-end learned representations, hand-crafted optical flow features are still widely used in video analysis tasks. To fill this gap, we propose TVNet, a novel end-to-end trainable neural network, to learn optical-flow-like features from data. TVNet subsumes a specific optical flow solver, the TV-L1 method, and is initialized by unfolding its optimization iterations as neural layers. TVNet can therefore be used directly without any extra learning. Moreover, it can be naturally concatenated with other task-specific networks to formulate an end-to-end architecture, thus making our method more efficient than current multi-stage approaches by avoiding the need to pre-compute and store features on disk. Finally, the parameters of the TVNet can be further fine-tuned by end-to-end training. This enables TVNet to learn richer and task-specific patterns beyond exact optical flow. Extensive experiments on two action recognition benchmarks verify the effectiveness of the proposed approach. Our TVNet achieves better accuracies than all compared methods, while being competitive with the fastest counterpart in terms of features extraction time.