Research map / 3D vision / Face and shape
Lifting 2D StyleGAN for 3D-Aware Face Generation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Distill pretrained StyleGAN2 with explicit 3D shape, pose, lighting, and differentiable rendering | StyleGAN2 face generation · FFHQ generated faces | FID: 12.90 (DiscoFaceGAN) vs 29.81 (LiftedGAN); 43.05 (CONFIGNet); 100.28 (HoloGAN) | Mixed |
| Add symmetric reconstruction loss (ablation) | LiftedGAN · StyleGAN2-generated faces | Depth error: 0.623 -> 0.435 | Helped |
| Add symmetric reconstruction loss (ablation) | LiftedGAN · StyleGAN2-generated faces | FID: 28.69 -> 29.81 | Hurt |
| Add perturbation loss (ablation) | LiftedGAN · StyleGAN2-generated faces | Depth error: 0.548 -> 0.435 | Helped |
| Add perturbation loss (ablation) | LiftedGAN · StyleGAN2-generated faces | FID: 21.30 -> 29.81 | Hurt |
| Add identity regularization loss (ablation) | LiftedGAN · StyleGAN2-generated faces | Depth error: 0.498 -> 0.435 | Helped |
| Add identity regularization loss (ablation) | LiftedGAN · StyleGAN2-generated faces | FID: 30.63 -> 29.81 | Helped |
| Add albedo consistency loss (ablation) | LiftedGAN · StyleGAN2-generated faces | Depth error: 0.467 -> 0.435 | Helped |
| Add albedo consistency loss (ablation) | LiftedGAN · StyleGAN2-generated faces | FID: 27.34 -> 29.81 | Hurt |
| Use 3D rendering rather than latent-only style manipulation for larger pose changes | LiftedGAN · Large face poses | Identity preservation | Helped |
Closest papers by meaning
- [Re] Lifting 2D StyleGAN for 3D-Aware Face Generation (NeurIPS 2022 · similarity 0.92)
- Dual Mapping of 2D StyleGAN for 3D-Aware Image Generation and Manipulation (Student Abstract) (AAAI 2024 · similarity 0.91)
- Self-Supervised Geometry-Aware Encoder for Style-Based 3D GAN Inversion (CVPR 2023 · similarity 0.90)
- In-N-Out: Faithful 3D GAN Inversion with Volumetric Decomposition for Face Editing (CVPR 2024 · similarity 0.87)
- Controllable 3D Face Generation with Conditional Style Code Diffusion (AAAI 2024 · similarity 0.87)
- Reference-Based 3D-Aware Image Editing with Triplanes (CVPR 2025 · 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.