Cross Modal Retrieval On Soundingearth
Metrics
Image-to-sound R@100
Median Rank
Sound-to-image R@100
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Image-to-sound R@100 | Median Rank | Sound-to-image R@100 |
|---|---|---|---|
| learning-tri-modal-embeddings-for-zero-shot | 0.434 | 159 | 0.434 |
| self-supervised-audiovisual-representation | 0.291 | 565 | 0.250 |