TSM+W3 (16 frames, ResNet50)

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 V1
Top 1 Accuracy· 2020-04-02
52.6
best: 70 (InternVideo)
Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention arXiv:2004.01278
Activity RecognitiononSomething-Something V1
Top 5 Accuracy· 2020-04-02
81.3
best: 91.9 (VideoMAE V2-g)
Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention arXiv:2004.01278

Time Series2 results

Action RecognitiononSomething-Something V1
Top 1 Accuracy· 2020-04-02
52.6
best: 70 (InternVideo)
Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention arXiv:2004.01278
Action RecognitiononSomething-Something V1
Top 5 Accuracy· 2020-04-02
81.3
best: 91.9 (VideoMAE V2-g)
Knowing What, Where and When to Look: Efficient Video Action Modeling with Attention arXiv:2004.01278