Counterfactual audio-image recognition on Ex-FlickrNet
73.67APC-MCR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| C-MCRSetting=Zero-shot2023.05 | 73.67 | 74.02 | |
| AudioCLIPSetting=Zero-shot2023.05 | 72.78 | 71.98 | |
| WAV2CLIPSetting=Zero-shot2023.05 | 60.54 | 66.2 | |
| SLAVCSetting=Zero-shot2023.05 | 52.15 | 60.1 | |
| EZ-VSLSetting=Zero-shot2023.05 | 48.75 | 56.8 | |
| OGLSetting=Zero-shot2023.05 | 40.2 | 55.7 | |
| DSOLSetting=Zero-shot2023.05 | 38.32 | 49.4 | |
| DMCSetting=Zero-shot2023.05 | 25.56 | 41.8 | |
| AttentionSetting=Zero-shot2023.05 | 15.98 | 24 |