M2A: Motion Aware Attention for Accurate Video Action Recognition

by   Brennan Gebotys, et al.

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 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 classes in the benchmark.


page 1

page 4

page 5


Few-shot Action Recognition via Improved Attention with Self-supervision

Most existing few-shot learning methods in computer vision focus on clas...

Where and When to Look? Spatio-temporal Attention for Action Recognition in Videos

Inspired by the observation that humans are able to process videos effic...

Mobile Video Action Recognition

Video action recognition, which is topical in computer vision and video ...

Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention

Attentive video modeling is essential for action recognition in unconstr...

TSI: Temporal Saliency Integration for Video Action Recognition

Efficient spatiotemporal modeling is an important yet challenging proble...

Paying More Attention to Motion: Attention Distillation for Learning Video Representations

We address the challenging problem of learning motion representations us...

CHAM: action recognition using convolutional hierarchical attention model

Recently, the soft attention mechanism, which was originally proposed in...

Please sign up or login with your details

Forgot password? Click here to reset