Diffusion Personalization Tuning Free On
평가 지표
Cosine Similarity
FID
LPIPS
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Cosine Similarity | FID | LPIPS | Paper Title | Repository |
|---|---|---|---|---|---|
| IP-Adapter-FaceID (SDXL) | 0.535 | 24.105 | 0.462 | IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models | |
| InstantID | 0.713 | 18.598 | 0.437 | InstantID: Zero-shot Identity-Preserving Generation in Seconds | - |
| IP-Adapter-FaceID-PlusV2 | 0.572 | 10.798 | 0.429 | IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models | |
| FastComposer | 0.256 | 31.736 | 0.487 | FastComposer: Tuning-Free Multi-Subject Image Generation with Localized Attention | |
| PhotoMaker | 0.287 | 8.410 | 0.424 | PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding | |
| Arc2Face | 0.796 | 6.628 | 0.508 | Arc2Face: A Foundation Model for ID-Consistent Human Faces | |
| IP-Adapter-FaceID-Plus | 0.600 | 11.817 | 0.384 | IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models |
0 of 7 row(s) selected.