中文版 | English

在结果中检索

CASIA OpenIR
（本次检索基于用户作品认领结果）

浏览/检索结果: 共59条，第1-10条 帮助

限定条件        
已选(0)清除 条数/页：   排序方式：
Understanding and Mitigating Overfitting in Prompt Tuning for Vision-Language Models 期刊论文
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2023, 卷号: 33, 期号: 9, 页码: 4616-4629
作者:  Ma, Chengcheng;  Liu, Yang;  Deng, Jiankang;  Xie, Lingxi;  Dong, Weiming;  Xu, Changsheng
Adobe PDF(1644Kb)  |  收藏  |  浏览/下载：78/11  |  提交时间：2023/11/16
Vision-language model  prompt tuning  over-fitting  subspace learning  gradient projection  
Recovering Generalization via Pre-training-like Knowledge Distillation for Out-of-Distribution Visual Question Answering 期刊论文
IEEE Transactions on Multimedia, 2023, 页码: 1-15
作者:  Song, Yaguang;  Yang, Xiaoshan;  Wang, Yaowei;  Xu, Changsheng
Adobe PDF(2397Kb)  |  收藏  |  浏览/下载：151/41  |  提交时间：2023/06/12
Multi-modal Foundation Model  Out-of-Distribution Generalization  Visual Question Answering  Knowledge Distillation  
Self-supervised Calorie-aware Heterogeneous Graph Networks for Food Recommendation 期刊论文
ACM Transactions on Multimedia Computing, Communications, and Applications, 2023, 卷号: 19, 期号: 1s, 页码: 1-23
作者:  Song, Yaguang;  Yang, Xiaoshan;  Xu, Changsheng
Adobe PDF(1381Kb)  |  收藏  |  浏览/下载：147/52  |  提交时间：2023/06/12
Food recommendation  recipe calories  heterogeneous graph  selfsupervised learning  
Exploring the Temporal Consistency of Arbitrary Style Transfer: A Channelwise Perspective 期刊论文
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 页码: 15
作者:  Kong, Xiaoyu;  Deng, Yingying;  Tang, Fan;  Dong, Weiming;  Ma, Chongyang;  Chen, Yongyong;  He, Zhenyu;  Xu, Changsheng
收藏  |  浏览/下载：122/0  |  提交时间：2023/03/20
Correlation  Task analysis  Optical imaging  Integrated optics  Lighting  Optical fiber networks  Image reconstruction  Arbitrary stylization  channel correlation  cross-domain  feature migration  
Weakly-supervised video object grounding via causal intervention 期刊论文
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023, 卷号: 45, 期号: 3, 页码: 3933 - 3948
作者:  Wang, Wei;  Gao, Junyu;  Xu, Changsheng
Adobe PDF(4558Kb)  |  收藏  |  浏览/下载：118/50  |  提交时间：2023/04/25
Relative Alignment Network for Source-Free Multimodal Video Domain Adaptation 会议论文
MM '22: Proceedings of the 30th ACM International Conference on Multimedia, Lisboa, Portugal, 2022.10.10—2022.10.14
作者:  Huang Yi;  Yang Xiaoshan;  Zhang Ji;  Xu Changsheng
Adobe PDF(1264Kb)  |  收藏  |  浏览/下载：155/65  |  提交时间：2023/06/21
Margin-Based Adversarial Joint Alignment Domain Adaptation 期刊论文
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2022, 卷号: 32, 期号: 4, 页码: 2057-2067
作者:  Zuo, Yukun;  Yao, Hantao;  Zhuang, Liansheng;  Xu, Changsheng
收藏  |  浏览/下载：237/0  |  提交时间：2022/06/10
Feature extraction  Adaptation models  Image reconstruction  Generative adversarial networks  Semisupervised learning  Data models  Training  Domain adaptation  joint alignment module  margin-based generative module  
Evo-ViT: Slow-Fast Token Evolution for Dynamic Vision Transformer 会议论文
, Online, 2022-3-1
作者:  Xu, Yifan;  Zhang, Zhijie;  Zhang, Mengdan;  Sheng, Kekai;  Li, Ke;  Dong, Weiming;  Zhang, Liqing;  Xu, Changsheng;  Sun, Xing
Adobe PDF(636Kb)  |  收藏  |  浏览/下载：235/63  |  提交时间：2022/06/14
Learning Hierarchical Video Graph Networks for One-Stop Video Delivery 期刊论文
ACM Transactions on Multimedia Computing, Communications, and Applications, 2022, 卷号: 18, 期号: 1, 页码: 1-23
作者:  Song, Yaguang;  Gao, Junyu;  Yang, Xiaoshan;  Xu, Changsheng
Adobe PDF(7608Kb)  |  收藏  |  浏览/下载：120/38  |  提交时间：2023/04/25
Cross modal  video retrieval  deep learning  graph neural networks  
Holographic Feature Learning of Egocentric-Exocentric Videos for Multi-Domain Action Recognition 期刊论文
IEEE TRANSACTIONS ON MULTIMEDIA, 2022, 卷号: 24, 页码: 2273-2286
作者:  Huang, Yi;  Yang, Xiaoshan;  Gao, Junyun;  Xu, Changsheng
Adobe PDF(2409Kb)  |  收藏  |  浏览/下载：287/61  |  提交时间：2022/07/25
Videos  Feature extraction  Visualization  Task analysis  Computational modeling  Target recognition  Prototypes  Egocentric videos  exocentric videos  holographic feature  multi-domain  action recognition