M2A: Motion Aware Attention for Accurate Video Action Recognition

TL;DR

This paper introduces Motion Aware Attention (M2A), a novel attention mechanism that explicitly incorporates motion information to improve video action recognition accuracy, outperforming existing methods with minimal added complexity.

Contribution

The paper proposes M2A, a simple, motion-aware attention mechanism that can be integrated into various neural networks to enhance video action recognition performance.

Findings

M2A improves top-1 accuracy by 15-26% across different architectures.

M2A outperforms other motion and attention mechanisms by up to 60% in specific classes.

Incorporating M2A yields significant gains on the Something-Something V1 benchmark.

Abstract

Advancements in attention mechanisms have led to significant performance improvements in a variety of areas in machine learning due to its ability to enable the dynamic modeling of temporal sequences. A particular area in computer vision that is likely to benefit greatly from the incorporation of attention mechanisms in video action recognition. However, much of the current research's focus on attention mechanisms have been on spatial and temporal attention, which are unable to take advantage of the inherent motion found in videos. Motivated by this, we develop a new attention mechanism called Motion Aware Attention (M2A) that explicitly incorporates motion characteristics. More specifically, M2A extracts motion information between consecutive frames and utilizes attention to focus on the motion patterns found across frames to accurately recognize actions in videos. The proposed M2A mechanism is simple to implement and can be easily incorporated into any neural network backbone architecture. We show that incorporating motion mechanisms with attention mechanisms using the proposed M2A mechanism can lead to a +15% to +26% improvement in top-1 accuracy across different backbone architectures, with only a small increase in computational complexity. We further compared the performance of M2A with other state-of-the-art motion and attention mechanisms on the Something-Something V1 video action recognition benchmark. Experimental results showed that M2A can lead to further improvements when combined with other temporal mechanisms and that it outperforms other motion-only or attention-only mechanisms by as much as +60% in top-1 accuracy for specific classes in the benchmark.