Action Segmentation on GTEA (F1 Thresholds, Edit, and Accuracy)
84AccuracyBIT
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| BITVideo Transcripts=true2023.08 | 84 | 96.1 | 95.6 | 87.5 | 96.3 | |
| UVASTVideo Transcripts=true2023.08 | 82.7 | 96.9 | 95.3 | 86 | — | |
| ASFormerVideo Transcripts=true2023.08 | 82 | 95.8 | 94.1 | 82.9 | — | |
| BIT2023.08 | 82 | 94.8 | 92.8 | 82.6 | 92.6 | |
| Br-Prompt+ASFormerFeatures=Bridge-Prompt (ViT), Backbone=ASFormer2022.03 | 81.2 | 94.1 | 92 | 83 | 91.6 | |
| C2F-TCN2022.05 | 80.8 | 90.3 | 88.8 | 77.7 | 86.4 | |
| CETNet2022.05 | 80.3 | 91.8 | 91.2 | 81.3 | 87.9 | |
| UVAST2023.08 | 80.2 | 92.7 | 91.3 | 81 | 92.1 | |
| MS-TCN++Backbone=I3D2022.03 | 80.1 | 88.8 | 85.7 | 76 | 83.5 | |
| MS-TCN++2022.05 | 80.1 | 88.8 | 85.7 | 76 | 83.5 | |
| MS-TCN++2023.08 | 80.1 | 88.8 | 85.7 | 76 | 83.5 | |
| BCN2021.10 | 79.8 | 88.5 | 87.1 | 77.3 | 84.4 | |
| BCNBackbone=I3D2022.03 | 79.8 | 88.5 | 87.1 | 77.3 | 84.4 | |
| SSTDABackbone=I3D2022.03 | 79.8 | 90 | 89.1 | 78 | 86.2 | |
| SSTDA2022.05 | 79.8 | 90 | 89.1 | 78 | 86.2 | |
| BCN2022.05 | 79.8 | 88.5 | 87.1 | 77.3 | 84.4 | |
| ASFormer2021.10 | 79.7 | 90.1 | 88.8 | 79.2 | 84.6 | |
| ASFormerFeatures=I3D2022.03 | 79.7 | 90.1 | 88.8 | 79.2 | 84.6 | |
| ASFormer2022.05 | 79.7 | 90.1 | 88.8 | 79.2 | 84.6 | |
| ASFormer2023.08 | 79.7 | 90.1 | 88.8 | 79.2 | 84.6 | |
| MS-TCN2022.05 | 79.2 | 87.5 | 85.4 | 74.6 | 81.4 | |
| SSTDA+HASRBackbone=I3D2022.03 | 78.7 | 90.9 | 88.6 | 76.4 | 87.5 | |
| SSTDA+HASR2022.05 | 78.7 | 90.9 | 88.6 | 76.4 | 87.5 | |
| Gao et al.2021.10 | 78.5 | 89.9 | 87.3 | 75.8 | 84.6 | |
| G2LBackbone=I3D2022.03 | 78.5 | 89.9 | 87.3 | 75.8 | 84.6 | |
| MSTCN++Video Transcripts=true2023.08 | 78.3 | 94.3 | 90.5 | 78.9 | — | |
| ETSN2022.05 | 78.2 | 91.1 | 90 | 77.9 | 86.2 | |
| DTGRM2021.10 | 77.6 | 87.8 | 86.6 | 72.9 | 83 | |
| ASRF2021.10 | 77.3 | 89.4 | 87.8 | 79.8 | 83.7 | |
| ASRFBackbone=I3D2022.03 | 77.3 | 89.4 | 87.8 | 79.8 | 83.7 | |
| ASRF2022.05 | 77.3 | 89.4 | 87.8 | 79.8 | 83.7 | |
| ASRF2023.08 | 77.3 | 89.4 | 87.8 | 79.8 | 83.7 | |
| MS-TCN2021.10 | 76.3 | 85.8 | 83.4 | 69.8 | 79 | |
| ASFormerFeatures=ViT2022.03 | 75.6 | 88.5 | 86.2 | 77.6 | 87.1 | |
| SSA-GAN2021.10 | 74.4 | 80.6 | 79.1 | 74.2 | 76 | |
| TDRN2021.10 | 70.1 | 79.2 | 74.4 | 62.7 | 74.1 | |
| ED-TCN2021.10 | 64 | 72.2 | 69.3 | 56 | — | |
| PL-StitchBackbone=ViT-B/16, Protocol=k-NN2025.11 | 62.4 | — | — | — | — | |
| ST-CNN2021.10 | 60.6 | 58.7 | 54.4 | 41.9 | — | |
| DINOBackbone=ViT-B/16, Protocol=k-NN2025.11 | 60 | — | — | — | — | |
| iBOTBackbone=ViT-B/16, Protocol=k-NN2025.11 | 59.2 | — | — | — | — | |
| Bi-LSTM2021.10 | 55.5 | 66.5 | 59 | 43.6 | — | |
| PL-StitchBackbone=ViT-B/16, Protocol=Linear Probing2025.11 | 54.1 | 61.7 | 50 | 30.4 | 60.2 | |
| DINOBackbone=ViT-B/16, Protocol=Linear Probing2025.11 | 52.2 | 55.3 | 46.7 | 28.3 | 53.9 | |
| iBOTBackbone=ViT-B/16, Protocol=Linear Probing2025.11 | 51.3 | 59.3 | 49 | 27.4 | 57.2 | |
| T-CoReBackbone=ViT-B/16, Protocol=Linear Probing2025.11 | 47.2 | 49.3 | 42.8 | 20.9 | 50 | |
| T-CoReBackbone=ViT-B/16, Protocol=k-NN2025.11 | 42.7 | — | — | — | — | |
| VideoMAEv2Backbone=ViT-B/16, Protocol=Linear Probing2025.11 | 30.2 | 23.5 | 17.8 | 10 | 22.8 | |
| VideoMAEv2Backbone=ViT-B/16, Protocol=k-NN2025.11 | 26.3 | — | — | — | — |