CSpace

浏览/检索结果: 共1条,第1-1条 帮助

已选(0)清除 条数/页:   排序方式:
Focus and Align: Learning Tube Tokens for Video-Language Pre-Training 期刊论文
IEEE TRANSACTIONS ON MULTIMEDIA, 2023, 卷号: 25, 页码: 8036-8050
作者:  Zhu, Yongqing;  Li, Xiangyang;  Zheng, Mao;  Yang, Jiahao;  Wang, Zihan;  Guo, Xiaoqian;  Chai, Zifeng;  Yuan, Yuchen;  Jiang, Shuqiang
收藏  |  浏览/下载:11/0  |  提交时间:2024/05/20
Electron tubes  Semantics  Visualization  Feature extraction  Task analysis  Transformers  Detectors  Local alignment mechanism  semantic centers  tube tokens  video-language pre-training