Image-to-audio Retrieval on MIAO (test)
61.45mAPProposed method
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Proposed methodDirection=I2A, Image Encoder=CLIP (frozen), Audio Encoder=CLAP (frozen), Projector=Two-layer MLP2026.05 | 61.45 | 53.6 | 68.9 | 60 | |
| BaselineDirection=I2A, Image Encoder=CLIP (ViT-B/32), Audio Encoder=CLAP (HTS-AT)2026.05 | 6.77 | 2 | 9 | 7.6 |