search

OpenGVLab / UniFormerV2

[ICCV2023] UniFormerV2: Spatiotemporal Learning by Arming Image ViTs with Video UniFormer
https://arxiv.org/abs/2211.09552
Apache License 2.0
294 stars 19 forks source link

feature extraction #8

Closed yhl2018 closed 1 year ago

yhl2018 commented 1 year ago

I want to use your method as extracted feature too！ could you help me? Modify the model to look like swin transformer or like MMAction2 backbone and cls head？

Andy1621 commented 1 year ago

Either is OK! It just depends on you. You may need to change the keys of state_dict if you modify the model.

You can follow https://github.com/OpenGVLab/UniFormerV2/issues/7. Just do it. Don't hesitate to ask me if you meet any problem.

Andy1621 commented 1 year ago

As there is no more activity, I am closing the issue, don't hesitate to reopen it if necessary.