Research map / 3D vision / Face and shape
ProgressiveAvatars: Progressive Animatable 3D Gaussian Avatars
Kaiwen Song, Jinkai Cui, Juyong Zhang
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 |
|---|---|---|---|
| supervise all levels (multi-level supervision) instead of finest only (ablation) | ProgressiveAvatars · NeRSemble NVS, 35% budget | PSNR: 20.06 -> 29.87 | Improved |
| supervise all levels (ablation) | ProgressiveAvatars · NeRSemble NES, 35% budget | PSNR: 19.86 -> 25.65 | Improved |
| importance-ordered transmission instead of random order (ablation) | ProgressiveAvatars · NeRSemble NVS, 25% budget | PSNR: 28.40 -> 29.14 | Improved |
| importance-ordered transmission instead of random order (ablation) | ProgressiveAvatars · NeRSemble NVS, 25% budget | LPIPS: 0.109 -> 0.080 | Improved |
| joint training of all levels instead of freezing previous levels (ablation) | ProgressiveAvatars · NeRSemble NVS, 100% | PSNR: 31.08 -> 31.47 | Improved |
| joint training of all levels instead of freezing previous levels (ablation) | ProgressiveAvatars · NeRSemble NES, 100% | LPIPS: 0.080 -> 0.080 | No clear effect |
| adaptive growing instead of uniform subdivision (ablation) | ProgressiveAvatars · NeRSemble | PSNR: 26.74 -> 27.89 | Improved |
| add multi-level hierarchy over importance ranking alone (ablation) | ProgressiveAvatars · NeRSemble, 50% budget | visual coverage | Improved |
| stream single progressive asset instead of discrete compression levels (ablation) | ProgressiveAvatars · NeRSemble | transmit data (MB): 227.2 -> 43.4 | Improved |
| progressive model at 100% vs GaussianAvatars | ProgressiveAvatars · NeRSemble NVS | PSNR: 31.10 -> 31.47 | Improved |
| progressive model at 100% vs GaussianAvatars | ProgressiveAvatars · NeRSemble NVS | SSIM: 0.937 -> 0.929 | Hurt |
| limit to 5% base budget | ProgressiveAvatars · NeRSemble NVS | PSNR: 31.47 -> 27.89 | 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.
- GaussianAvatars — GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (CVPR 2024) · better on 2 of 3 shared results
- PointAvatar — PointAvatar: Deformable Point-Based Head Avatars From Videos (CVPR 2023) · better on 2 of 2 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
- FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding (CVPR 2024 · similarity 0.89)
- GASP: Gaussian Avatars with Synthetic Priors (CVPR 2025 · similarity 0.89)
- GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning (CVPR 2024 · similarity 0.87)
- RGBAvatar: Reduced Gaussian Blendshapes for Online Modeling of Head Avatars (CVPR 2025 · similarity 0.87)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.87)
- GPAvatar: High-fidelity Head Avatars by Learning Efficient Gaussian Projections (CVPR 2025 · 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.
