Research map / 3D vision / Face and shape
Sequential Gaussian Avatars with Hierarchical Motion Context
Wangze Xu, Yifan Zhan, Zhihang Zhong, Xiao Sun
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 vanilla non-rigid MLP with pose condition (ablation) | SeqAvatar baseline · I3D-Human | PSNR: 29.76 -> 31.05 | Improved |
| add skeleton motion condition (ablation) | SeqAvatar · I3D-Human | PSNR: 31.05 -> 31.89 | Improved |
| add point-wise velocity condition (ablation) | SeqAvatar · I3D-Human | PSNR: 31.89 -> 32.01 | Improved |
| add spatio-temporal multi-scale sampling (ablation) | SeqAvatar · I3D-Human | LPIPS*: 31.23 -> 29.78 | Improved |
| add spatio-temporal multi-scale sampling (ablation) | SeqAvatar · I3D-Human | PSNR: 32.01 -> 32.24 | Improved |
| use three sampling steps instead of one (ablation) | SeqAvatar · I3D-Human | PSNR: 32.11 -> 32.24 | Improved |
| use 8 KNN vertexes instead of 1 (ablation) | SeqAvatar · I3D-Human | PSNR: 32.14 -> 32.24 | Improved |
| use 10 KNN vertexes instead of 8 (ablation) | SeqAvatar · I3D-Human | PSNR: 32.24 -> 32.20 | No clear effect |
| use SeqAvatar instead of Dyco | Dyco · ZJU-MoCap | LPIPS*: 29.47 -> 28.89 | Improved |
| use SeqAvatar instead of GauHuman | GauHuman · ZJU-MoCap | PSNR: 31.04 -> 31.02 | No clear effect |
| use SeqAvatar instead of 3DGS-Avatar | 3DGS-Avatar · ZJU-MoCap | LPIPS*: 27.09 -> 28.89 | Hurt |
| use SeqAvatar instead of Dyco | Dyco · I3D-Human novel pose | LPIPS*: 35.32 -> 36.07 | 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.
- 3DGS-Avatar — 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024) · better on 4 of 5 shared results
- GauHuman — GauHuman: Articulated Gaussian Splatting from Monocular Human Videos (CVPR 2024) · better on 3 of 4 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
- MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics (NeurIPS 2025 · similarity 0.90)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.90)
- GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (CVPR 2024 · similarity 0.89)
- GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning (CVPR 2024 · similarity 0.89)
- SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting (CVPR 2024 · similarity 0.88)
- GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (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.
