search

PKU-YuanGroup / Chat-UniVi

[CVPR 2024 Highlight🔥] Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding
https://arxiv.org/abs/2311.08046
Apache License 2.0
755 stars 41 forks source link

About the visualization #7

Closed pd162 closed 9 months ago

pd162 commented 9 months ago

Thanks for your outstanding work! And I have a question about the visualization of paper. I noticed that dynamic visual tokens of Fig. 1 in the paper, and I also find corresponding core funtion vis_token in TCFormer, but I cannot reproduce this picture. I want to ask to visualize dynamic tokens in detail. It would be even better if visualization scripts could be open-source!

jpthu17 commented 9 months ago

I have released the visualization script. You can locate it in VISUALIZATION.md.

pd162 commented 9 months ago

Thx! Looks great!