- 博客(3)
- 收藏
- 关注
原创 Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained 论文笔记
Vision-language models (VLMs) pre-trained on largescale image-text pairs have demonstrated impressive transferability on various visual tasks. Transferring knowledge from such powerful VLMs is a promising direction for building effective video recognition
2025-03-17 12:00:51
1789
1
空空如也
空空如也
TA创建的收藏夹 TA关注的收藏夹
TA关注的人