Vocal Tract Segmentation on 75-Speaker US-ST Annot-16
87.74DSC (%)Ours
Evaluation results
| Method | Links | ||
|---|---|---|---|
| OursMethod Type=Multimodal2026.05 | 87.74 | 2.11 | |
| Jain et al.Method Type=Multimodal2026.05 | 84.68 | 2.5 | |
| AV-SAMMethod Type=Multimodal, Adapted=true2026.05 | 83.47 | 3.01 | |
| AVSegFormerMethod Type=Multimodal, Adapted=true2026.05 | 81.27 | 2.39 | |
| U-MambaMethod Type=Unimodal2026.05 | 80.9 | 2.98 | |
| DINOv2Method Type=Unimodal2026.05 | 77.7 | 3.51 | |
| MedSAM-2Method Type=Unimodal2026.05 | 73.29 | 3.17 | |
| nnU-NetMethod Type=Unimodal2026.05 | 73.11 | 5.9 | |
| Swin-UNETRMethod Type=Unimodal2026.05 | 72.72 | 5.06 | |
| ResUNet-aMethod Type=Unimodal2026.05 | 68.27 | 6.6 | |
| MedSegDiff-V2Method Type=Unimodal2026.05 | 67.04 | 5.08 |