Self-supervised segmentation, after the paper hype
SimCLR and MoCo pretraining gave me a ResNet50 that worked. Here is what I would do differently if I were starting over today.
For my final-year project I pretrained a ResNet50 with SimCLR and MoCo on a corpus of medical images, then fine-tuned it for segmentation against a fully supervised baseline. The self-supervised encoder won on the small labelled set, lost on the large one, and the crossover point was somewhere around 2,000 labelled images.Read the post




