Sound-to-Image Retrieval on SoundingEarth 0.2m GSD (test)
43.4Recall@100GeoCLAP
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP, Text Modality=true, Address Modality=true2023.09 | 43.4 | 167 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP, Text Modality=false, Address Modality=true2023.09 | 43.1 | 167 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP, Text Modality=true, Address Modality=false2023.09 | 41.9 | 175 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP, Text Modality=false, Address Modality=false2023.09 | 38.5 | 237 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP (frozen), Text Modality=false, Address Modality=false2023.09 | 34.8 | 369 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP (frozen), Text Modality=true, Address Modality=false2023.09 | 32.5 | 428 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP (frozen), Text Modality=true, Address Modality=true2023.09 | 31.1 | 443 | |
| GeoCLAPImage Encoder=SATMAE, Text-Audio Encoder=L-CLAP (frozen), Text Modality=false, Address Modality=true2023.09 | 29.5 | 544 | |
| BaselineImage Encoder=ResNet18, Text-Audio Encoder=ResNet18, Text Modality=false, Address Modality=false2023.09 | 25 | 816 |