search

facebookresearch / dinov2

PyTorch code and models for the DINOv2 self-supervised learning method.
Apache License 2.0
9.03k stars 791 forks source link

dinov2 for mobilenet or mobilevit version please #138

Closed TranThanh96 closed 1 year ago

TranThanh96 commented 1 year ago

as the title, can you guys try this ? I think that is good if we have a light weight version for mobile

qasfb commented 1 year ago

The issue with this is that (from my understanding) the mobilenet / mobilevit models usually have a pyramid-like architecture, with poolings, so it is not immediately easy to perform distillation on the patch features because the output maps have different dimensionality; however if you have an efficient architecture in mind that would output e.g. 16x16 feature maps + a class token, we could consider it.

patricklabatut commented 1 year ago

Closing to keep track of similar asks in #166 instead

TranThanh96 commented 1 year ago

The issue with this is that (from my understanding) the mobilenet / mobilevit models usually have a pyramid-like architecture, with poolings, so it is not immediately easy to perform distillation on the patch features because the output maps have different dimensionality; however if you have an efficient architecture in mind that would output e.g. 16x16 feature maps + a class token, we could consider it.

how about this? https://github.com/snap-research/EfficientFormer https://github.com/apple/ml-fastvit