Cross-view Action Recognition Understanding From Exocentric to Egocentric Perspective

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Truong, Thanh-Dat, Luu, Khoa
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916369067409408
author Truong, Thanh-Dat
Luu, Khoa
author_facet Truong, Thanh-Dat
Luu, Khoa
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15699
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Cross-view Action Recognition Understanding From Exocentric to Egocentric Perspective
Truong, Thanh-Dat
Luu, Khoa
Computer Vision and Pattern Recognition
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.
title Cross-view Action Recognition Understanding From Exocentric to Egocentric Perspective
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.15699