ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders

Open paper
Benchmark Model Rank Results
image-classification-on-places365ViC-MAE (ViT-L)#3Top 1 Accuracy: 59.5%
action-recognition-in-videos-on-something-something-v2ViC-MAE (ViT-L)#19Top-1 Accuracy: 73.7
action-classification-on-kinetics-400ViC-MAE (ViT-L)#52Acc@1: 85.1
image-classification-on-imagenetViC-MAE (ViT-L)#287Top 1 Accuracy: 85%

Rank counts only results with a code link.