View-independent gait events detection using CNN-transformer hybrid network

Ankhzaya Jamsrandorj, Dawoon Jung, Konki Sravan Kumar, Muhammad Zeeshan Arshad, Hwasup Lim, Jinwook Kim, Kyung-Ryoul Mun

Abstract: Highlights•A reliable vision-based approach for gait event detection is proposed.•The proposed method achieved outstanding performance in detecting both standard and peculiar gaits.•The gait events were accurately detected regardless of camera positions or angles and walking speed conditions.•The proposed approach could offer significant practical benefits and convenience.