Research map / 3D vision / Face and shape
MeshAvatar: Learning High-quality Triangular Human Avatars from Multi-view Videos
Yushuo Chen, Zerong Zheng, Zhe Li, Chao Xu, Yebin Liu
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 |
|---|---|---|---|
| use hybrid mesh-plus-implicit representation with radiance rendering instead of prior avatar methods | MeshAvatar (radiance) · AvatarReX training frame reconstruction | PSNR: 28.4015 -> 28.9939 | Improved |
| replace radiance rendering with physics-based rendering (ablation) | MeshAvatar · AvatarReX novel pose synthesis | PSNR: 26.6194 -> 26.8526 PSNR; LPIPS 0.0657 -> 0.0693 | Mixed |
| add pseudo normal supervision (ablation) | MeshAvatar · SyntheticHuman++ | Normal error (degrees): 19.37 -> 13.44 | Improved |
| add pseudo normal supervision (ablation) | MeshAvatar · SyntheticHuman++ relighting | PSNR: 26.92 -> 27.19 | Improved |
| add pseudo normal supervision (ablation) | MeshAvatar · SyntheticHuman++ visibility | PSNR: 20.36 -> 22.30 | Improved |
| combine shadow-aware PBR with pseudo normal supervision (ablation) | MeshAvatar · training frame geometry (qualitative) | geometry quality | Improved |
| use MeshAvatar without normal prior against AvatarReX | MeshAvatar (w/o normal) · AvatarReX novel pose synthesis | LPIPS / SSIM: LPIPS 0.0684 -> 0.0687 | No clear effect |
| compare MeshAvatar normal accuracy to Xu et al. on synthetic data | MeshAvatar · SyntheticHuman++ | Normal error (degrees): 12.44 -> 13.44 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Relighting4D — Relighting4D: Neural Relightable Human from Videos (ECCV 2022) · better on 2 of 2 shared results
- ARAH — ARAH: Animatable Volume Rendering of Articulated Human SDFs (ECCV 2022) · better on 1 of 1 shared result
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
- GETAvatar: Generative Textured Meshes for Animatable Human Avatars (ICCV 2023 · similarity 0.89)
- Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025 · similarity 0.89)
- Neural Image-based Avatars: Generalizable Radiance Fields for Human Avatar Modeling (ICLR 2023 · similarity 0.89)
- GeoAvatar: Geometrically-Consistent Multi-Person Avatar Reconstruction from Sparse Multi-View Videos (CVPR 2025 · similarity 0.88)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.88)
- GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning (CVPR 2024 · 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.
