Multi-source sound localization on VGGSound Instruments (test)
89.6CIoU@0.1T-VSL
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| T-VSLEncoders=AudioCLIP (ResNet-50 + ESResNeXt)2024.04 | 89.6 | 41.8 | 31.5 | |
| AVGNEncoders=AudioCLIP pre-trained2024.04 | 86.7 | 38.1 | 28.8 | |
| DSOLEncoders=AudioCLIP pre-trained2024.04 | 85.9 | — | 28.4 | |
| AlignmentEncoders=AudioCLIP pre-trained2024.04 | 84.9 | 35.6 | 27.1 | |
| Mix-and-LocalizeEncoders=AudioCLIP pre-trained2024.04 | 84.5 | 36.3 | 26.7 | |
| FNACEncoders=AudioCLIP pre-trained2024.04 | 84.1 | 35.4 | 26.8 | |
| AudioCLIPEncoders=AudioCLIP pre-trained2024.04 | 83.8 | 34.7 | 25.9 | |
| MarginNCEEncoders=AudioCLIP pre-trained2024.04 | 82.3 | 33.5 | 24.7 | |
| LVSEncoders=AudioCLIP pre-trained2024.04 | 81.1 | — | 23.2 | |
| EZ-VSLEncoders=AudioCLIP pre-trained2024.04 | 80.3 | — | 22.9 | |
| Wav2CLIPEncoders=AudioCLIP pre-trained2024.04 | 80.3 | 31.2 | 22.2 | |
| CoarsetoFineEncoders=AudioCLIP pre-trained2024.04 | 78.8 | — | 19.7 | |
| OTSEncoders=AudioCLIP pre-trained2024.04 | 77.8 | 28.7 | 18.2 |