标题：Gait Recognition with A Single Sample per Person
作者：Li, Wei; Peng, Jingliang
作者机构：[Li, Wei; Peng, Jingliang] Shandong Univ, Sch Comp Sci & Technol, Jinan, Peoples R China.
会议名称：Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)
会议日期：DEC 13-16, 2016
来源：2016 ASIA-PACIFIC SIGNAL AND INFORMATION PROCESSING ASSOCIATION ANNUAL SUMMIT AND CONFERENCE (APSIPA)
摘要：Gait recognition with a single sample per person (SSPP) is a challenging problem but has so far drawn little research attention. Inspired by similar research in face recognition, we propose to utilize the intra-class variation information learned from an additional generic training set with multiple samples per person to improve the representation of the query sample. We learn a sparse variation dictionary from the generic training set by connecting the generic training set with the gallery set adaptively. The sparse variation dictionary is then integrated into the framework of sparse representation based classification so that various variations in clothing, object carrying and so forth can be better addressed. Experiments on the publicly available CASIA gait database demonstrate the promising performance of the proposed scheme for gait recognition with SSPP.