ViC-MAE (ViT-L)

Reported on 3 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision1 result

VideoonKinetics-400
Acc@1· 2023-03-21
85.1
best: 93.6 (OmniVec2)
ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders arXiv:2303.12001

Robots1 result

Activity RecognitiononSomething-Something V2
Top-1 Accuracy· 2023-03-21
73.7
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders arXiv:2303.12001

Time Series1 result

Action RecognitiononSomething-Something V2
Top-1 Accuracy· 2023-03-21
73.7
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
ViC-MAE: Self-Supervised Representation Learning from Images and Video with Contrastive Masked Autoencoders arXiv:2303.12001