Research map / 3D vision / Human pose estimation
Pose-Guided Human Animation From a Single Image in the Wild
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Separate silhouette prediction from the rendering network (ablation) | SGR · Human pose transfer | synthesis quality | Helped |
| Add predicted target garment labels (ablation) | SGR · Human pose transfer | boundary quality | Helped |
| Remove source garment labels (ablation) | SGR · Human pose transfer | performance | Hurt |
| Condition RenderNet on the source style code (ablation) | SGR · Human pose transfer | generation quality | Helped |
| Condition on pseudo target images warped from coherent UV maps (ablation) | SGR · Human pose transfer | appearance preservation | Helped |
| Use multiple input views to complete coherent UV maps (ablation) | SGR · Human pose transfer | clothing texture preservation | Helped |
| Remove the source silhouette mask (ablation) | SGR · Human pose transfer | synthesis quality | Mixed |
| Remove the source garment label map (ablation) | SGR · Human pose transfer | synthesis quality | Hurt |
| Remove the source spatial style code (ablation) | SGR · Human pose transfer | generation quality | Hurt |
| Remove the RenderNet KL loss and use fully connected layers for the latent space (ablation) | SGR · Human pose transfer | synthesis quality | Mixed |
| Replace the completed UV texture map with direct source-image warping (ablation) | SGR · Human pose transfer | appearance preservation | Hurt |
| Remove the RenderNet module and its intermediate outputs (ablation) | SGR · Human pose transfer | synthesis quality | Mixed |
Closest papers by meaning
- Free-viewpoint Human Animation with Pose-correlated Reference Selection (CVPR 2025 · similarity 0.90)
- DisPose: Disentangling Pose Guidance for Controllable Human Image Animation (ICLR 2025 · similarity 0.88)
- HumanNeRF-SE: A Simple yet Effective Approach to Animate HumanNeRF with Diverse Poses (CVPR 2024 · similarity 0.88)
- HumanVid: Demystifying Training Data for Camera-controllable Human Image Animation (NeurIPS 2024 · similarity 0.88)
- ReMask-Animate: Refined Character Image Animation Using Mask-Guided Adapters (AAAI 2025 · similarity 0.88)
- MagicPose: Realistic Human Poses and Facial Expressions Retargeting with Identity-aware Diffusion (ICML 2024 · similarity 0.87)
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.