satseg

For MIE1520 at UofT

Much of the code in here has been borrowed/adapted from:

• https://sampl-weizmann.github.io/DeepCut/
• https://github.com/ShirAmir/dino-vit-features
• https://github.com/huggingface/pytorch-image-models/blob/main/timm/models/vision_transformer.py

To run:

1. Clone repo
2. pip install -r requirements_bare.txt
3. Available scripts:
   • segment.py to generate image segmentations (-h for help)
   • correspondence.py to generate correspondence maps
   • attention_maps.py to generate... well...