Audio-visual localization on VGGSound (Unheard 110 categories)
48.4cIoUSOUPLE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SOUPLEObject-Guided Refinement (OGL)=false2026.03 | 48.4 | 46.24 | |
| SOUPLEObject-Guided Refinement (OGL)=false, Training Epochs=502026.03 | 48.4 | 46.24 | |
| AlignmentObject-Guided Refinement (OGL)=true2026.03 | 42.94 | 41.54 | |
| FNACObject-Guided Refinement (OGL)=true2026.03 | 42.91 | 41.17 | |
| ACL-SSLObject-Guided Refinement (OGL)=false2026.03 | 41.98 | 41.55 | |
| EZ-VSLObject-Guided Refinement (OGL)=true2026.03 | 39.57 | 39.6 | |
| AlignmentObject-Guided Refinement (OGL)=false2026.03 | 39.11 | 39.8 | |
| SLAVCObject-Guided Refinement (OGL)=true2026.03 | 38.87 | — | |
| SLAVCObject-Guided Refinement (OGL)=false2026.03 | 36.5 | — | |
| EZ-VSLObject-Guided Refinement (OGL)=false2026.03 | 32.66 | 36.72 | |
| LVS2026.03 | 26.3 | 34.7 |