CASIA OpenIR
(本次检索基于用户作品认领结果)

浏览/检索结果: 共7条,第1-7条 帮助

限定条件        
已选(0)清除 条数/页:   排序方式:
VLP: A Survey on Vision-language Pre-training 期刊论文
Machine Intelligence Research, 2023, 卷号: 20, 期号: 1, 页码: 38-56
作者:  Fei-Long Chen;  Du-Zhen Zhang;  Ming-Lun Han;  Xiu-Yi Chen;  Jing Shi;  Shuang Xu;  Bo Xu
Adobe PDF(1427Kb)  |  收藏  |  浏览/下载:55/17  |  提交时间:2024/04/23
Vision and language  pre-training  transformers  multimodal learning  representation learning  
VLP: A Survey on Vision-language Pre-training 期刊论文
Machine Intelligence Research, 2023, 卷号: 20, 期号: 1, 页码: 38-56
作者:  Feilong Chen;  Duzhen Zhang;  Minglun Han;  Xiuyi Chen;  Jing Shi;  Shuang Xu;  Bo Xu
Adobe PDF(969Kb)  |  收藏  |  浏览/下载:176/34  |  提交时间:2023/06/21
DMRM: A Dual-Channel Multi-Hop Reasoning Model for Visual Dialog 会议论文
, 美国纽约, 2020.2
作者:  Feilong Chen;  Fandong Meng;  Jiaming Xu;  Peng Li;  Bo Xu;  Jie Zhou
Adobe PDF(3052Kb)  |  收藏  |  浏览/下载:148/35  |  提交时间:2023/06/07
IMPROVING CROSS-MODAL UNDERSTANDING IN VISUAL DIALOG VIA CONTRASTIVE LEARNING 会议论文
, Singapore, 2022.5
作者:  Feilong Chen;  Duzhen Zhang;  Xiuyi Chen;  Jing Shi;  Shuang Xu;  Bo Xu
Adobe PDF(9035Kb)  |  收藏  |  浏览/下载:240/102  |  提交时间:2023/06/07
Unsupervised and Pseudo-Supervised Vision-Language Alignment in Visual Dialog 会议论文
, Lisboa, Portugal, October 10–14, 2022
作者:  Feilong Chen;  Duzhen Zhang;  Xiuyi Chen;  Jing Shi;  Shang Xu;  Bo Xu
Adobe PDF(9035Kb)  |  收藏  |  浏览/下载:284/157  |  提交时间:2023/06/05
End-to-End Chinese Image Text Recognition with Attention Model 会议论文
, Guangzhou, China, 2017-11-14 ~ 2017-11-18
作者:  Sheng, Fenfen;  Zhai, Chuanlei;  Chen, Zhineng;  Xu, Bo
浏览  |  Adobe PDF(1061Kb)  |  收藏  |  浏览/下载:253/81  |  提交时间:2020/06/12
Pyrboxes: An efficient multi-scale scene text detector with feature pyramids 期刊论文
PATTERN RECOGNITION LETTERS, 2019, 卷号: 125, 期号: 2019, 页码: 228-234
作者:  Sheng, Fenfen;  Chen, Zhineng;  Zhang, Wei;  Xu, Bo
浏览  |  Adobe PDF(1558Kb)  |  收藏  |  浏览/下载:360/60  |  提交时间:2019/12/16
Scene text detection  Multi-scale text detection  Grouped pyramid module  Efficient and effective