Cross-view Action Recognition Understanding From Exocentric to Egocentric Perspective

TL;DR

This paper introduces a novel cross-view learning method for action recognition that transfers knowledge from large-scale exocentric videos to egocentric videos using geometric constraints and a new self-attention loss, achieving state-of-the-art results.

Contribution

It proposes a geometric-based constraint integrated into Transformer self-attention and a cross-view self-attention loss for effective knowledge transfer across views.

Findings

Achieves state-of-the-art performance on Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100.

Demonstrates the effectiveness of geometric constraints in cross-view attention.

Shows improved transfer learning from exocentric to egocentric videos.

Abstract

Understanding action recognition in egocentric videos has emerged as a vital research topic with numerous practical applications. With the limitation in the scale of egocentric data collection, learning robust deep learning-based action recognition models remains difficult. Transferring knowledge learned from the large-scale exocentric data to the egocentric data is challenging due to the difference in videos across views. Our work introduces a novel cross-view learning approach to action recognition (CVAR) that effectively transfers knowledge from the exocentric to the selfish view. First, we present a novel geometric-based constraint into the self-attention mechanism in Transformer based on analyzing the camera positions between two views. Then, we propose a new cross-view self-attention loss learned on unpaired cross-view data to enforce the self-attention mechanism learning to transfer knowledge across views. Finally, to further improve the performance of our cross-view learning approach, we present the metrics to measure the correlations in videos and attention maps effectively. Experimental results on standard egocentric action recognition benchmarks, i.e., Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100, have shown our approach's effectiveness and state-of-the-art performance.