Lee, J. Kim, C. Park, S.-K. 2024-01-19T20:04:14Z 2024-01-19T20:04:14Z 2021-08-31 2019-05 1976-5622 https://pubs.kist.re.kr/handle/201004/120077 Most current gait recognition approaches based on convolution neural networks (CNNs) do not learn the discriminative features of separable inter-class differences resulting from cross-view data. To improve this discriminative ability, this paper proposes a network that reduces intra-class variation using a center loss function for view-invariant gait recognition. The proposed method achieved 92% accuracy using OU-MVLP, the largest existing gait recognition dataset. Furthermore, a network trained using the OU-MVLP achieved 95% accuracy with the OU-LP. These results demonstrate that the proposed method offers a good generalization performance. ？ ICROS 2019. Korean Institute of Control, Robotics and Systems Convolution Gait analysis Convolution neural network Discriminative ability Discriminative features Gait energy images Gait identifications Gait recognition Generalization performance Intra-class variation Pattern recognition Cross-view gait identification based on convolution neural network with joint loss function Article 10.5302/J.ICROS.2019.19.0066 1 Journal of Institute of Control, Robotics and Systems, v.25, no.5, pp.463 - 469 Journal of Institute of Control, Robotics and Systems 25 5 463 469 scopus kci ART002463894 2-s2.0-85066797193 Article Convolution Gait analysis Convolution neural network Discriminative ability Discriminative features Gait energy images Gait identifications Gait recognition Generalization performance Intra-class variation Pattern recognition Center loss Convolution neural network Discriminative-feature learning Gait energy image Gait recognition