Fig. 7From: Multi-view large population gait dataset and its performance evaluation for cross-view gait recognitionNetwork architectures of CNN-based methods. Numbers written on left side of conv, pool, fc indicate [(#kernel) ×(kernel size, width × height) / (stride)], [(#kernel size, width × height) / (stride)], [(#output node)] respectively. N s written on 1in-GEINet indicates the number of training subjects. a 1in-GEINet, bSiamese, c MT, d LBBack to article page