Cross-view Action Recognition Understanding From Exocentric to Egocentric Perspective

arXiv:2305.15699v2 Announce Type: replace
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 …

abstract action action recognition applications arxiv collection cs.cv data data collection deep learning knowledge perspective practical recognition replace research robust scale type understanding videos view vital