Extract oxford features and resize transform

facebookresearch / vissl

VISSL is FAIR's library of extensible, modular and scalable components for SOTA Self-Supervised Learning with images.

https://vissl.ai

MIT License

3.26k stars 334 forks source link

Closed iseessel closed 3 years ago

iseessel commented 3 years ago

Summary:

Setup oxford feature extraction so that features can be reused in instance_retrieval_test.
Create ImgPilResize to resize LONGER side to 1024. Torchvision.Resize can only transform smaller side.

Differential Revision: D30179807

facebook-github-bot commented 3 years ago

This pull request was exported from Phabricator. Differential Revision: D30179807

facebook-github-bot commented 3 years ago

This pull request was exported from Phabricator. Differential Revision: D30179807

facebook-github-bot commented 3 years ago

This pull request was exported from Phabricator. Differential Revision: D30179807