Joint Audio-Video Generation on Custom human-centric audio-video real and anime images (test)
8.15PQOmniAvatar
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| OmniAvatarGeneration Paradigm=Two-stage Generation, Parameters (S+V)=21.1B2025.11 | 8.15 | 7.41 | 0.152 | 0.987 | 0 | 0.721 | 6.34 | 0.454 | 0.349 | |
| Wan-S2VGeneration Paradigm=Two-stage Generation, Parameters (S+V)=16.6B2025.11 | 8.15 | 7.41 | 0.152 | 0.991 | 0.13 | 0.75 | 6.35 | 0.481 | 0.375 | |
| UniAVGenGeneration Paradigm=Joint Generation, Joint Training Samples=1.3M, Parameters (S+V)=7.1B2025.11 | 7 | 6.62 | 0.151 | 0.973 | 0.41 | 0.779 | 5.95 | 0.832 | 0.573 | |
| OviGeneration Paradigm=Joint Generation, Joint Training Samples=30.7M, Parameters (S+V)=10.9B2025.11 | 6.03 | 6.01 | 0.216 | 0.972 | 0.36 | 0.774 | 6.48 | 0.828 | 0.558 | |
| JavisDiTGeneration Paradigm=Joint Generation, Joint Training Samples=10.1M, Parameters (S+V)=3.7B2025.11 | 5.21 | 3.93 | 0.986 | 0.965 | 0.373 | 0.716 | 1.23 | 0.776 | 0.388 | |
| Universe-1Generation Paradigm=Joint Generation, Joint Training Samples=6.4M, Parameters (S+V)=7.1B2025.11 | 4.56 | 4.29 | 0.296 | 0.985 | 0.08 | 0.733 | 1.21 | 0.573 | 0.3 |