Research map / Video / Video generation
OmniFace: Bridging the Image-to-Video Gap for High-Fidelity Face Swapping via Diffusion Transformer
Xu Guo, Fulong Ye, Xinghui Li, Pengqi Tu, Pengze Zhang, Qichao Sun, Songtao Zhao, Xiangwang Hou, Qian HE
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add SyncID-Pipe ID quadruplet supervision instead of self-reconstruction inpainting (ablation) | OmniFace · IDBench-V | ID-Arc: 0.510 -> 0.659 | Improved |
| add SyncID-Pipe ID quadruplet supervision instead of self-reconstruction inpainting (ablation) | OmniFace · IDBench-V | Variance: 0.0036 -> 0.0029 | Hurt |
| add synthetic GT data with real GT data (Synthetic-to-Real Training) (ablation) | OmniFace · IDBench-V | ID-Arc: 0.604 -> 0.659 | Improved |
| add synthetic GT data with real GT data (Synthetic-to-Real Training) (ablation) | OmniFace · IDBench-V | VFID: 2.145 -> 2.243 | Mixed |
| add real paired data to forward-generated paired data (ablation) | OmniFace · IDBench-V | VFID: 3.845 -> 2.243 | Improved |
| add real paired data to forward-generated paired data (ablation) | OmniFace · IDBench-V | ID-Arc: 0.657 -> 0.659 | No clear effect |
| add Identity-Coherence Reward Weighting (IDRW) (ablation) | OmniFace · IDBench-V | ID-Arc: 0.631 -> 0.659 | Improved |
| add Identity-Coherence Reward Weighting (IDRW) (ablation) | OmniFace · IDBench-V | Variance: 0.0041 -> 0.0029 | Improved |
| add Identity-Coherence Reward Weighting (IDRW) (ablation) | OmniFace · IDBench-V | Pose error: 2.687 -> 2.446 | Improved |
| OmniFace vs DreamID image swapping applied per frame | OmniFace · IDBench-V | ID-Arc: 0.616 -> 0.659 | Improved |
| OmniFace vs CanonSwap on pose error | OmniFace · IDBench-V | Pose error: 2.430 -> 2.446 | Hurt |
| OmniFace vs DreamID in user study | OmniFace · User study (19 volunteers) | Quality (1-5): 3.06 -> 4.15 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CanonSwap — CanonSwap: High-Fidelity and Consistent Video Face Swapping via Canonical Space Modulation (ICCV 2025) · better on 3 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- VividFace: A Robost and High-Fidelity Video Face Swapping Framework (NeurIPS 2025 · similarity 0.93)
- DynamicFace: High-Quality and Consistent Face Swapping for Image and Video using Composable 3D Facial Priors (ICCV 2025 · similarity 0.91)
- Preserving Source Video Realism: High-Fidelity Face Swapping for Cinematic Quality (CVPR 2026 · similarity 0.91)
- CanonSwap: High-Fidelity and Consistent Video Face Swapping via Canonical Space Modulation (ICCV 2025 · similarity 0.91)
- StableAnimator: High-Quality Identity-Preserving Human Image Animation (CVPR 2025 · similarity 0.88)
- FaithfulFaces: Pose-Faithful Facial Identity Preservation for Text-to-Video Generation (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
