Audio-visual localization on VGGSound (Heard 110 categories)
54.85cIoUSOUPLE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SOUPLEObject-Guided Refinement (OGL)=false, Training Epochs=502026.03 | 54.85 | 48.98 | |
| SOUPLEObject-Guided Refinement (OGL)=false2026.03 | 54.76 | 48.86 | |
| ACL-SSLObject-Guided Refinement (OGL)=false2026.03 | 48.44 | 45.06 | |
| AlignmentObject-Guided Refinement (OGL)=true2026.03 | 41.85 | 40.93 | |
| FNACObject-Guided Refinement (OGL)=true2026.03 | 39.54 | 39.83 | |
| AlignmentObject-Guided Refinement (OGL)=false2026.03 | 38.31 | 39.05 | |
| SLAVCObject-Guided Refinement (OGL)=true2026.03 | 38.22 | — | |
| EZ-VSLObject-Guided Refinement (OGL)=true2026.03 | 37.25 | 38.97 | |
| SLAVCObject-Guided Refinement (OGL)=false2026.03 | 35.84 | — | |
| EZ-VSLObject-Guided Refinement (OGL)=false2026.03 | 31.86 | 36.19 | |
| LVS2026.03 | 28.9 | 36.2 |