[RFC] New Augmentation techniques in Torchvison

🚀 Feature

Inclusion of new Augmentation techniques in torchvision.transforms.

Motivation

Transforms are important for data augmentation 😅

Proposals

• RandAugment Citations 306 #4348
• Cutout Citations 964 Superseded by RandomErasing
• MixUp Citations 1624 Code (Note it is CC-BY-NC-license) #4379
• CutMix Citations 437 Code ICCV 2019 #4379
• TrivialAugment ICCV 2021 #4221
• AugMix Citations 157
• Scale-Jitter Citations 11 - code - benchmarks
• Simple CopyPaste #5825
• Port SSD & SSDlite Augmentations and Mixup/Cutmix from references to vision
• Detection Transforms #1406 #2213 #3980
• AutoAugment Detection 231 Citations code
• Greedy Search Policy
• FastAutoAugment Citations 135 NeurIPS 2019 Code (1k+ GitHub stars)
• ReMixMatch Citations 165
• FixMatch Citations 258
• Transforms for color spaces augmentations E.g. rgb2hsv, rgb2bgr, rgb2lab, etc #4029
• Gaussian Noise See #6192

Additional context

To visitors
Kindly give a 👍 if you think any of these would help in your work.

Also if you have any transform in mind please provide few details here!

Linked to #3221

cc @vfdev-5 @fmassa

Start with the open proposals in torchvision.transforms and read the linked references/detection/transforms.py and references/classification/transforms.py entry points, along with related issues #1406, #2213, #3980, #4029, and #6192. First choose and scope one remaining augmentation rather than treating the full RFC as a single task. Done means the selected proposal is implemented in vision and its checklist item can be resolved.