CelebV-HQ
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Talking Head Generation | CelebV-HQ | FID0 | 15 | |
| Talking Head Generation | CelebV-HQ (test) | FID6.818 | 11 | |
| Talking Avatar Generation | CelebV-HQ (clips) | FID37.54 | 10 | |
| Portrait Image Animation | CelebV-HQ identity-disjoint (test) | FID7.61 | 8 | |
| Video self-reconstruction | CelebV-HQ (test) | L1 Error0.0376 | 8 | |
| Talking Head Synthesis | CelebV-HQ (test) | IQA3.588 | 6 | |
| Facial Attribute Recognition | CelebV-HQ | Appearance Accuracy93.9 | 6 | |
| Identity-Preserving Talking Head Generation | CelebV-HQ hard (cross-video) | Speaker Similarity47.7 | 5 | |
| Identity-Preserving Talking Head Generation | CelebV-HQ easy (same-video) | Spk Sim0.573 | 5 | |
| Audio Generation | CelebV-HQ | FAD244.83 | 5 | |
| Unconditional Video Generation | CelebV-HQ 256^2 | FVD (16 frames)56.1 | 5 | |
| Image-conditioned text-to-speech | CelebV-HQ | MCD-DTW7.9383 | 2 | |
| Face Rendering | CelebV-HQ (test) | L1 Loss0.003 | 2 | |
| Face Anonymization | CelebV-HQ | ID Preservation0.011 | 2 | |
| Facial Expression Classification | CelebV-HQ (test) | Metric- | 0 |
Showing 15 of 15 rows