Research map / 3D vision / Face and shape
Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields
Keqiang Sun, Shangzhe Wu, Zhaoyang Huang, Ning Zhang, Quan Wang, Hongsheng Li
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 mesh-guided volume sampler (MgS) (ablation) | pi-GAN + L_recon baseline · CelebA | Chamfer distance: 0.87 -> 0.29 | Improved |
| add mesh-guided volume sampler (MgS) (ablation) | pi-GAN + L_recon baseline · CelebA | FID (128): 21.97 -> 38.91 | Hurt |
| add distance-aware density regularizer Rd (ablation) | +MgS model · CelebA | Landmark correlation: 27.45 -> 51.74 | Improved |
| add 3D landmark loss (ablation) | cGOF · CelebA | Landmark correlation: 31.63 FID row: LC 30.25 -> 84.43 | Improved |
| add volume warping loss (ablation) | cGOF · CelebA | Landmark correlation: 84.43 -> 89.91 | Improved |
| add 3D landmark loss (ablation) | cGOF · CelebA | FID (128): 31.63 -> 56.90 | Hurt |
| add smoothness regularizer on depth maps (ablation) | cGOF · CelebA | Landmark correlation: 89.91 -> 92.88 | Improved |
| add direct depth loss between NeRF and 3DMM depth (ablation) | pi-GAN + L_recon baseline · CelebA | Landmark correlation: 26.15 -> 0.06 | Hurt |
| add neural renderer (64 to 512 resolution) (ablation) | cGOF · CelebA | FID (512): 71.60 -> 31.83 | Improved |
| add neural renderer (ablation) | cGOF · CelebA | DS expression: 29.13 -> 27.47 | Hurt |
| use cGOF instead of 2D 3DMM-based generator (DiscoFaceGAN) | controllable face synthesis · CelebA | DS expression: 15.70 -> 27.47 | Improved |
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.
- HeadNeRF — HeadNeRF: A Real-Time NeRF-Based Parametric Head Model (CVPR 2022) · better on 2 of 2 shared results
- StyleRig — StyleRig: Rigging StyleGAN for 3D Control Over Portrait Images (CVPR 2020) · 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
- Controllable 3D Face Generation with Conditional Style Code Diffusion (AAAI 2024 · similarity 0.87)
- Text-Guided 3D Face Synthesis - From Generation to Editing (CVPR 2024 · similarity 0.87)
- Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation (CVPR 2024 · similarity 0.87)
- Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusion (CVPR 2025 · similarity 0.86)
- DiffPortrait3D: Controllable Diffusion for Zero-Shot Portrait View Synthesis (CVPR 2024 · similarity 0.86)
- Text-conditional Attribute Alignment across Latent Spaces for 3D Controllable Face Image Synthesis (CVPR 2024 · similarity 0.86)
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.
