Research map / Video / Video segmentation and tracking
Learning Motion Refinement for Unsupervised Face Animation
Jiale Tao, Shuhang Gu, Wen Li, Lixin Duan
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 iterative motion refinement to the prior-based model (ablation) | Motion Transformer · Voxceleb1 | AKD: 1.190 -> 1.155 | Improved |
| Add iterative motion refinement to the prior-based model | Motion Transformer · CelebV-HQ | AKD | Improved |
| Use non-prior-based initialization without a prior motion model (ablation) | NPMR · Voxceleb1 | AKD: 1.190 -> 1.328 | Hurt |
| Add NPMR to FOMM (ablation) | FOMM · Voxceleb1 | AKD: 1.254 -> 1.224 | Improved |
| Add NPMR to TPSM (ablation) | TPSM · Voxceleb1 | AKD: 1.231 -> 1.176 | Improved |
| Add NPMR to MTIA (ablation) | MTIA · Voxceleb1 | AKD: 1.190 -> 1.155 | Improved |
| Add NPMR to FOMM with 10 keypoints (ablation) | FOMM-kp10 · Voxceleb1 | AKD: 1.254 -> 1.224 | Improved |
| Add NPMR to FOMM with 15 keypoints (ablation) | FOMM-kp15 · Voxceleb1 | AKD: 1.245 -> 1.199 | Improved |
| Add NPMR to FOMM with 20 keypoints (ablation) | FOMM-kp20 · Voxceleb1 | AKD: 1.222 -> 1.194 | Improved |
| Remove motion-flow updates and refine occlusion only (ablation) | Ours full · Voxceleb1 | AKD: 1.155 -> 1.172 | Hurt |
| Remove occlusion updates and refine flow only (ablation) | Ours full · Voxceleb1 | AKD: 1.155 -> 1.175 | Hurt |
| Remove warped source features from the flow updater (ablation) | Ours full · Voxceleb1 | AKD: 1.155 -> 1.160 | Hurt |
Closest papers by meaning
- X-NeMo: Expressive Neural Motion Reenactment via Disentangled Latent Attention (ICLR 2025 · similarity 0.86)
- Synergizing Motion and Appearance: Multi-Scale Compensatory Codebooks for Talking Head Video Generation (CVPR 2025 · similarity 0.86)
- RealPortrait: Realistic Portrait Animation with Diffusion Transformers (AAAI 2025 · similarity 0.86)
- FSRT: Facial Scene Representation Transformer for Face Reenactment from Factorized Appearance Head-pose and Facial Expression Features (CVPR 2024 · similarity 0.86)
- KeyFace: Expressive Audio-Driven Facial Animation for Long Sequences via KeyFrame Interpolation (CVPR 2025 · similarity 0.86)
- PTUS: Photo-Realistic Talking Upper-Body Synthesis via 3D-Aware Motion Decomposition Warping (AAAI 2024 · similarity 0.86)
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.
