Self-supervised masked autoencoders for surgical phase recognition
Self-supervised Pretrained Models:
- MAE
- MaskFeat
- VideoMae
- DINO
Backbone Models:
- MViTv2
- Swin
- ViT
Phase Recognition Models:
- TMRNet
- Trans-SVNet
Saved weights for Trans-SVNet can be downloaded from: https://drive.google.com/drive/folders/1Zgrg1G8fdrOzQ5W3FREl35NmDQMDWEXp
Work done at the Surgical Artificial Intelligence & Innovation Laboratory (SAIIL), in collaboration with MIT.