MaskFeat (Kinetics-600 pretrain, MViT-L)

Reported on 2 benchmarks across 2 tasks · 1 paper · 2 SOTA

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

Robots1 result

Activity RecognitiononAVA v2.2
mAP· uses extra data· 2021-12-16
39.8
best: 45.1 (LART (Hiera-H, K700 PT+FT))
SOTA
Masked Feature Prediction for Self-Supervised Visual Pre-Training arXiv:2112.09133

Time Series1 result

Action RecognitiononAVA v2.2
mAP· uses extra data· 2021-12-16
39.8
best: 45.1 (LART (Hiera-H, K700 PT+FT))
SOTA
Masked Feature Prediction for Self-Supervised Visual Pre-Training arXiv:2112.09133