CASIA OpenIR
(本次检索基于用户作品认领结果)

浏览/检索结果: 共6条,第1-6条 帮助

限定条件                            
已选(0)清除 条数/页:   排序方式:
Extremely Lightweight Skeleton-Based Action Recognition With ShiftGCN plus 期刊论文
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2021, 卷号: 30, 页码: 7333-7348
作者:  Cheng, Ke;  Zhang, Yifan;  He, Xiangyu;  Cheng, Jian;  Lu, Hanqing
Adobe PDF(3205Kb)  |  收藏  |  浏览/下载:259/14  |  提交时间:2021/11/03
Skeleton-based action recognition  graph convolutional network  lightweight network  shift network  
Skeleton-Based Action Recognition with Shift Graph Convolutional Network 会议论文
, 线上, June 2020
作者:  Ke Cheng;  Yifan Zhang;  Xiangyu He;  Weihan Chen;  Jian Cheng;  Hanqing Lu
Adobe PDF(2935Kb)  |  收藏  |  浏览/下载:236/39  |  提交时间:2021/05/28
Rethinking the pid optimizer for stochastic optimization of deep networks 会议论文
, London, United kingdom, July 6, 2020 - July 10, 2020
作者:  Shi, Lei;  Zhang, Yifan;  Wang, Wanguo;  Cheng, Jian;  Lu, Hanqing
浏览  |  Adobe PDF(325Kb)  |  收藏  |  浏览/下载:219/55  |  提交时间:2021/01/27
Two-Stream Adaptive Graph Convolutional Networks for Skeleton-Based Action Recognition 会议论文
, Long Beach, CA, United states, June 16, 2019 - June 20, 2019
作者:  Shi, Lei;  Zhang, Yifan;  Cheng, Jian;  Lu, Hanqing
浏览  |  Adobe PDF(691Kb)  |  收藏  |  浏览/下载:228/59  |  提交时间:2021/01/26
Skeleton-Based Action Recognition With Multi-Stream Adaptive Graph Convolutional Networks 期刊论文
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2020, 期号: 29, 页码: 9532-9545
作者:  Shi, Lei;  Zhang, Yifan;  Cheng, Jian;  Lu, Hanqing
浏览  |  Adobe PDF(2849Kb)  |  收藏  |  浏览/下载:377/148  |  提交时间:2020/11/05
Skeleton-based action recognition, graph convolutional network, adaptive graph, multi-stream network.  
Gesture recognition based on deep deformable 3D convolutional neural networks 期刊论文
PATTERN RECOGNITION, 2020, 期号: 107, 页码: 12
作者:  Zhang, Yifan;  Shi, Lei;  Wu, Yi;  Cheng, Ke;  Cheng, Jian;  Lu, Hanqing
浏览  |  Adobe PDF(1310Kb)  |  收藏  |  浏览/下载:453/132  |  提交时间:2020/08/31
Gesture recognition  Spatiotemporal deformable convolution  Spatiotemporal convolutional neural network