CAST: Cross-Attention in Space and Time for Video Action Recognition
This work addresses the need for balanced spatio-temporal understanding in video action recognition, which is an incremental improvement over existing methods.
The paper tackles the problem of recognizing human actions in videos by proposing a two-stream architecture called CAST, which uses cross-attention to balance spatial and temporal understanding, achieving consistent performance across multiple benchmarks like EPIC-KITCHENS-100, Something-Something-V2, and Kinetics-400.
Recognizing human actions in videos requires spatial and temporal understanding. Most existing action recognition models lack a balanced spatio-temporal understanding of videos. In this work, we propose a novel two-stream architecture, called Cross-Attention in Space and Time (CAST), that achieves a balanced spatio-temporal understanding of videos using only RGB input. Our proposed bottleneck cross-attention mechanism enables the spatial and temporal expert models to exchange information and make synergistic predictions, leading to improved performance. We validate the proposed method with extensive experiments on public benchmarks with different characteristics: EPIC-KITCHENS-100, Something-Something-V2, and Kinetics-400. Our method consistently shows favorable performance across these datasets, while the performance of existing methods fluctuates depending on the dataset characteristics.