Research map / 3D vision / Face and shape
DipGuava: Disentangling Personalized Gaussian Features for 3D Head Avatars from Monocular Video
Jeonghaeng Lee, Seok Keun Choi, Zhixuan Li, Weisi Lin, Sanghoon Lee
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 staged training instead of joint fine-tuning of base and residual (ablation) | DipGuava · 18 monocular videos | PSNR: 29.02 -> 30.14 | Improved |
| train base at full resolution instead of 256^2 then 512^2 (ablation) | DipGuava · 18 monocular videos | LPIPS: 0.0949 -> 0.0417 | Improved |
| use surface normal input instead of optimizing base appearance (ablation) | DipGuava · 18 monocular videos | PSNR: 28.70 -> 30.14 | Improved |
| add adaptive UV sampling (ablation) | DipGuava · 18 monocular videos | PSNR: 29.53 -> 30.14 | Improved |
| use U-Net backbone instead of MLP (ablation) | DipGuava · 18 monocular videos | PSNR: 28.32 -> 30.14 | Improved |
| add residual branch R on top of base B (ablation) | DipGuava · 18 monocular videos | PSNR: 29.54 -> 30.14 | Improved |
| add base stage B before residual R (ablation) | DipGuava · 18 monocular videos | PSNR: 28.25 -> 30.14 | Improved |
| add residual geometric deformation delta G (ablation) | DipGuava · 18 monocular videos | PSNR: 29.94 -> 30.14 | Improved |
| add residual appearance R in two-stage model (ablation) | DipGuava · 18 monocular videos | PSNR: 29.95 -> 30.14 | Improved |
| add dynamic appearance fusion (ablation) | DipGuava · 18 monocular videos | PSNR: 29.86 -> 30.14 | Improved |
| add dynamic appearance fusion (ablation) | DipGuava · 18 monocular videos | LPIPS: 0.0410 vs 0.0417 | No clear effect |
| add dynamic appearance fusion (ablation) | DipGuava · 18 monocular videos | L1: 0.0091 -> 0.0090 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- FlashAvatar — FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding (CVPR 2024) · better on 4 of 4 shared results
- GaussianAvatars — GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians (CVPR 2024) · better on 4 of 4 shared results
- SplattingAvatar — SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting (CVPR 2024) · better on 4 of 4 shared results
- PointAvatar — PointAvatar: Deformable Point-Based Head Avatars From Videos (CVPR 2023) · better on 4 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
- Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians (CVPR 2024 · similarity 0.90)
- Gaussian Eigen Models for Human Heads (CVPR 2025 · similarity 0.90)
- Synthetic Prior for Few-Shot Drivable Head Avatar Inversion (CVPR 2025 · similarity 0.90)
- MeGA: Hybrid Mesh-Gaussian Head Avatar for High-Fidelity Rendering and Head Editing (CVPR 2025 · similarity 0.90)
- HRAvatar: High-Quality and Relightable Gaussian Head Avatar (CVPR 2025 · similarity 0.89)
- 3D Gaussian Head Avatars with Expressive Dynamic Appearances by Compact Tensorial Representations (CVPR 2025 · similarity 0.89)
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.
