search

microsoft / esvit

EsViT: Efficient self-supervised Vision Transformers
MIT License
409 stars 46 forks source link

Allow arbitrary image sizes and upstream changes from Swin-Transformer-Object-Detection #17

Open vadimkantorov opened 2 years ago

vadimkantorov commented 2 years ago

It is useful in object detection context to allow arbitrary sizes by doing dynamic mask computation (probably possible only with relative position encoding).

These kinds of edits were done in https://github.com/SwinTransformer/Swin-Transformer-Object-Detection and in https://github.com/megvii-research/SOLQ/. It would be nice if you upstreamed these changes. This will simplify trying out ESviT checkpoints as pretraining for object detection.

Also, fyi I created a similar issue in SimMIM: https://github.com/microsoft/SimMIM/issues/13. Overall, having some stable version of swin_transformer.py somewhere (maybe even in main SwinTransformer/Swin-Transformer repo?) supporting dynamic masking would help a lot :)

Thanks!

sym0926 commented 3 months ago

Hi,do you have ckpt and train logs , can you share with me ? I got an error ,when I download them.