Phase recognition on AutoLaparo (test)
70.7F1 ScoreAutoLaparo SSL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| AutoLaparo SSLBackbone=ResNet-50, Number of Labeled Videos=102023.12 | 70.7 | — | |
| ImageNet SupervisedBackbone=ResNet-50, Number of Labeled Videos=102023.12 | 70.1 | — | |
| EndoFMBackbone=ViT-B, Number of Labeled Videos=102023.12 | 68.9 | — | |
| ImageNet SupervisedBackbone=ViT-B, Number of Labeled Videos=102023.12 | 68.6 | — | |
| AutoLaparo SSLBackbone=ResNet-50, Number of Labeled Videos=52023.12 | 64.4 | — | |
| ImageNet SupervisedBackbone=ViT-B, Number of Labeled Videos=52023.12 | 62.3 | — | |
| Cholec80 SSLBackbone=ResNet-50, Number of Labeled Videos=102023.12 | 61.5 | — | |
| AutoLaparo SSLBackbone=ResNet-50, Number of Labeled Videos=32023.12 | 59.5 | — | |
| EndoFMBackbone=ViT-B, Number of Labeled Videos=52023.12 | 59.3 | — | |
| ImageNet SupervisedBackbone=ViT-B, Number of Labeled Videos=32023.12 | 58.5 | — | |
| Cholec80 SSLBackbone=ResNet-50, Number of Labeled Videos=52023.12 | 58.3 | — | |
| ImageNet SupervisedBackbone=ResNet-50, Number of Labeled Videos=52023.12 | 57.6 | — | |
| EndoFMBackbone=ViT-B, Number of Labeled Videos=32023.12 | 56.6 | — | |
| ImageNet SupervisedBackbone=ResNet-50, Number of Labeled Videos=32023.12 | 50.8 | — | |
| Cholec80 SSLBackbone=ResNet-50, Number of Labeled Videos=32023.12 | 47.7 | — | |
| HyperVLPPre-training Dataset=SVL, Zero-shot=true2026.06 | 32.9 | 42.9 | |
| PeskaVLPPre-training Dataset=SVL, Evaluation Protocol=Zero-shot2024.09 | 23.6 | 26.5 | |
| PeskaVLPPre-training Dataset=SVL, Zero-shot=true2026.06 | 23.6 | 26.5 | |
| HecVLPre-training Dataset=SVL, Evaluation Protocol=Zero-shot2024.09 | 18.9 | 23.3 | |
| HecVLPre-training Dataset=SVL, Zero-shot=true2026.06 | 18.9 | 23.3 | |
| SurgVLPPre-training Dataset=SVL, Evaluation Protocol=Zero-shot2024.09 | 16.6 | 21.3 | |
| SurgVLPPre-training Dataset=SVL, Zero-shot=true2026.06 | 16.6 | 21.3 | |
| CLIPPre-training Dataset=SVL, Evaluation Protocol=Zero-shot2024.09 | 16.2 | 18.9 | |
| CLIPPre-training Dataset=SVL, Zero-shot=true2026.06 | 16.2 | 18.9 | |
| CLIPPre-training Dataset=Scratch, Evaluation Protocol=Zero-shot2024.09 | 10.9 | 15.3 | |
| CLIPPre-training Dataset=Scratch, Zero-shot=true2026.06 | 10.9 | 15.3 | |
| CLIPPre-training Dataset=CLIP400M, Evaluation Protocol=Zero-shot2024.09 | 9.1 | 17.4 | |
| CLIPPre-training Dataset=CLIP400M, Zero-shot=true2026.06 | 9.1 | 17.4 | |
| MIL-NCEPre-training Dataset=Howto100M, Evaluation Protocol=Zero-shot2024.09 | 7.9 | 9.9 | |
| MIL-NCEPre-training Dataset=HowTo100M, Zero-shot=true2026.06 | 7.9 | 9.9 |