BEVT (IN-1K + Kinetics400 pretrain)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Robots2 results

Activity RecognitiononSomething-Something V2
Parameters· uses extra data· 2021-12-02
89
best: 2131 (InternVideo2-6B)
BEVT: BERT Pretraining of Video Transformers arXiv:2112.01529
Activity RecognitiononSomething-Something V2
Top-1 Accuracy· uses extra data· 2021-12-02
71.4
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
BEVT: BERT Pretraining of Video Transformers arXiv:2112.01529

Time Series2 results

Action RecognitiononSomething-Something V2
Parameters· uses extra data· 2021-12-02
89
best: 2131 (InternVideo2-6B)
BEVT: BERT Pretraining of Video Transformers arXiv:2112.01529
Action RecognitiononSomething-Something V2
Top-1 Accuracy· uses extra data· 2021-12-02
71.4
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
BEVT: BERT Pretraining of Video Transformers arXiv:2112.01529