CASIA OpenIR

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:
Cascade learning from adversarial synthetic images for accurate pupil detection 期刊论文
PATTERN RECOGNITION, 2019, 卷号: 88, 期号: 2020, 页码: 584-594
作者:  Gou, Chao;  Zhang, Hui;  Wang, Kunfeng;  Wang, Fei-Yue;  Ji, Qiang
浏览  |  Adobe PDF(2057Kb)  |  收藏  |  浏览/下载:400/113  |  提交时间:2019/07/12
Cascade regression  GANs  Pupil detection  
Part-aligned pose-guided recurrent network for action recognition 期刊论文
PATTERN RECOGNITION, 2019, 卷号: 92, 期号: 1, 页码: 165-176
作者:  Huang, Linjiang;  Huang, Yan;  Ouyang, Wanli;  Wang, Liang
浏览  |  Adobe PDF(2555Kb)  |  收藏  |  浏览/下载:526/119  |  提交时间:2019/07/11
Action recognition  Part alignment  Auto-transformer attention  
Efficient conic fitting with an analytical Polar-N-Direction geometric distance 期刊论文
PATTERN RECOGNITION, 2019, 卷号: 90, 页码: 415-423
作者:  Wu, Yihong;  Wang, Haoren;  Tang, Fulin;  Wang, Zhiheng
浏览  |  Adobe PDF(2113Kb)  |  收藏  |  浏览/下载:418/155  |  提交时间:2019/05/06
Conic fitting  Geometric distance  Sampson distance  
MAPNet: Multi-modal attentive pooling network for RGB-D indoor scene classification 期刊论文
PATTERN RECOGNITION, 2019, 期号: 90, 页码: 436-449
作者:  Li, Yabei;  Zhang, Zhang;  Cheng, Yanhua;  Wang, Liang;  Tan, Tieniu
浏览  |  Adobe PDF(5797Kb)  |  收藏  |  浏览/下载:393/38  |  提交时间:2019/04/23
Indoor scene classification  Multi-modal fusion  RGB-D  Attentive pooling  
Asymmetric 3D Convolutional Neural Networks for action recognition 期刊论文
PATTERN RECOGNITION, 2019, 卷号: 85, 期号: 1, 页码: 1-12
作者:  Yang, Hao;  Yuan, Chunfeng;  Li, Bing;  Du, Yang;  Xing, Junliang;  Hu, Weiming;  Maybank, Stephen J.
浏览  |  Adobe PDF(2689Kb)  |  收藏  |  浏览/下载:534/113  |  提交时间:2019/01/08
Asymmetric 3D convolution  MicroNets  3D-CNN  Action recognition  
Pseudo low rank video representation 期刊论文
PATTERN RECOGNITION, 2019, 卷号: 85, 期号: 1, 页码: 50-59
作者:  Yu, Tingzhao;  Wang, Lingfeng;  Guo, Chaoxu;  Gu, Huxiang;  Xiang, Shiming;  Pan, Chunhong
浏览  |  Adobe PDF(1456Kb)  |  收藏  |  浏览/下载:601/180  |  提交时间:2019/01/08
Pseudo low rank  Data driven  Low resolution  Action recognition