Context Autoencoder for Self-Supervised Representation Learning

Benchmark Model Rank Results
self-supervised-image-classification-on-imagenet-finetunedCAE (ViT-L/16)#15Top 1 Accuracy: 86.3%Number of Params: 307M
semantic-segmentation-on-ade20kCAE (ViT-L, UperNet)#57Validation mIoU: 54.7
object-detection-on-coco-val2017CAE* (ViT-L, 1600ep, Mask R-CNN 1x)#118box AP: 54.5AP50: 75.2AP75: 60.1