Audio-to-image Retrieval on MIAO (test)
61.08mAPProposed method
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Proposed methodDirection=A2I, Image Encoder=CLIP (frozen), Audio Encoder=CLAP (frozen), Projector=Two-layer MLP2026.05 | 61.08 | 64.6 | 88.2 | 0.75 | |
| BaselineDirection=A2I, Image Encoder=CLIP (ViT-B/32), Audio Encoder=CLAP (HTS-AT)2026.05 | 7.82 | 6 | 10 | 0.116 |