摘要: Image pre-training, the current de-facto paradigm for a wide range of visual tasks, is generally less favored in the field of video recognition. By contrast, a common strategy is to directly train with spatiotemporal convolutional... 展开
作者 | Xianhang Li Huiyu Wang Chen Wei Jieru Mei Alan Yuille Yuyin Zhou Cihang Xie | ||
---|---|---|---|
作者单位 | |||
文集名称 | Computer Vision – ECCV 2022 | ||
出版年 | 2022 | ||
会议名称 | European Conference on Computer Vision | ||
卷/页码 | Part 25 / 675-691 | 开始页/总页数 | 00000675 / 17 |
会议地点 | Tel Aviv(IL) | 会议年/会议届次 | 2022 / 17th |
关键词 | Video classification Imagenet pre-training 3d convolution networks | ||
馆藏号 | P2300511 |