DiffAct

Reported on 32 benchmarks across 2 tasks · 1 paper · 12 SOTA

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

Computer Vision32 results

Action Localizationon50 Salads
Acc· 2023-03-31
88.9
best: 91.4 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
Acc· 2023-03-31
76.4
best: 78 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
Average F1· 2023-03-31
73.6
best: 76.2 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
F1@10%· 2023-03-31
80.3
best: 82.1 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
F1@25%· 2023-03-31
75.9
best: 79 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
F1@50%· 2023-03-31
64.6
best: 67.5 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action Segmentationon50 Salads
Acc· 2023-03-31
88.9
best: 91.4 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
Acc· 2023-03-31
76.4
best: 78 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
Average F1· 2023-03-31
73.6
best: 76.2 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
F1@10%· 2023-03-31
80.3
best: 82.1 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
F1@25%· 2023-03-31
75.9
best: 79 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
F1@50%· 2023-03-31
64.6
best: 67.5 (AdaFocus (newly extracted I3D-features, LT-Context model))
SOTA
Diffusion Action Segmentation arXiv:2303.17959
Action Localizationon50 Salads
Edit· 2023-03-31
85
best: 89.1 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action Localizationon50 Salads
F1@10%· 2023-03-31
90.1
best: 92.7 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action Localizationon50 Salads
F1@25%· 2023-03-31
89.2
best: 91.6 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action Localizationon50 Salads
F1@50%· 2023-03-31
83.7
best: 88.5 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonGTEA
Acc· 2023-03-31
82.2
best: 89.8 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonGTEA
Edit· 2023-03-31
89.6
best: 93.5 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonGTEA
F1@10%· 2023-03-31
92.5
best: 96.1 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonGTEA
F1@25%· 2023-03-31
91.5
best: 95.6 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonGTEA
F1@50%· 2023-03-31
84.7
best: 91.3 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action LocalizationonBreakfast
Edit· 2023-03-31
78.4
best: 79.7 (FACT (efficient hybrid of convolution and transformer model))
Diffusion Action Segmentation arXiv:2303.17959
Action Segmentationon50 Salads
Edit· 2023-03-31
85
best: 89.1 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action Segmentationon50 Salads
F1@10%· 2023-03-31
90.1
best: 92.7 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action Segmentationon50 Salads
F1@25%· 2023-03-31
89.2
best: 91.6 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action Segmentationon50 Salads
F1@50%· 2023-03-31
83.7
best: 88.5 (Br-Prompt+ASPnet (RGB, flow, accelerometer))
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonGTEA
Acc· 2023-03-31
82.2
best: 89.8 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonGTEA
Edit· 2023-03-31
89.6
best: 93.5 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonGTEA
F1@10%· 2023-03-31
92.5
best: 96.1 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonGTEA
F1@25%· 2023-03-31
91.5
best: 95.6 (FACT)
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonGTEA
F1@50%· 2023-03-31
84.7
best: 91.3 (Semantic2Graph)
Diffusion Action Segmentation arXiv:2303.17959
Action SegmentationonBreakfast
Edit· 2023-03-31
78.4
best: 79.7 (FACT (efficient hybrid of convolution and transformer model))
Diffusion Action Segmentation arXiv:2303.17959