CASIA OpenIR

浏览/检索结果: 共5条,第1-5条 帮助

限定条件        
已选(0)清除 条数/页:   排序方式:
Weakly Aligned Feature Fusion for Multimodal Object Detection 期刊论文
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2021, 页码: 15
作者:  Zhang, Lu;  Liu, Zhiyong;  Zhu, Xiangyu;  Song, Zhan;  Yang, Xu;  Lei, Zhen;  Qiao, Hong
Adobe PDF(19222Kb)  |  收藏  |  浏览/下载:235/5  |  提交时间:2022/01/27
Object detection  Feature extraction  Detectors  Robustness  Cameras  Automation  Training  Deep learning  feature fusion  multimodal object detection  pedestrian detection  
Immobilizing Caging Grasps of Convex Polyhedrons With a Four-Pin Gripper 期刊论文
IEEE ROBOTICS AND AUTOMATION LETTERS, 2021, 卷号: 6, 期号: 4, 页码: 7683-7690
作者:  Su, Jianhua;  Liu, Chuankai;  Meng, Yan
Adobe PDF(2462Kb)  |  收藏  |  浏览/下载:311/48  |  提交时间:2021/11/03
Grasping  Grippers  Pins  Three-dimensional displays  Industries  Shape  Neck  Caging grasps  grasping  in-hand manipulation  industrial robot  
Adaptive Deep Metric Learning for Affective Image Retrieval and Classification 期刊论文
IEEE TRANSACTIONS ON MULTIMEDIA, 2021, 卷号: 23, 页码: 1640-1653
作者:  Yao, Xingxu;  She, Dongyu;  Zhang, Haiwei;  Yang, Jufeng;  Cheng, Ming-Ming;  Wang, Liang
收藏  |  浏览/下载:247/0  |  提交时间:2021/08/15
Measurement  Visualization  Semantics  Feature extraction  Task analysis  Image analysis  Image retrieval  Affective image retrieval  convolutional neural network  deep metric learning  visual sentiment analysis  
Visual affordance detection using an efficient attention convolutional neural network 期刊论文
NEUROCOMPUTING, 2021, 卷号: 440, 期号: 2021, 页码: 36-44
作者:  Gu, Qipeng;  Su, Jianhua;  Yuan, Lei
Adobe PDF(1561Kb)  |  收藏  |  浏览/下载:355/71  |  提交时间:2021/06/15
Affordance detection  Attention mechanism  Up-sampling layer  
Sparse self-attention aggregation networks for neural sequence slice interpolation 期刊论文
BioData Mining, 2021, 卷号: 14, 期号: 1, 页码: 19
作者:  Wang, Zejin;  Liu, Jing;  Chen, Xi;  Li, Guoqing;  Han, Hua
Adobe PDF(5022Kb)  |  收藏  |  浏览/下载:311/44  |  提交时间:2021/02/18
Slice interpolation  Biological tissue recovery  EM images  Sparse self-attention network  Adaptive style-balance loss