Research map / 3D vision / NeRF and radiance fields
StyleNeRF: A Style-based 3D Aware Generator for High-resolution Image Synthesis
Read the paperCheck 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 |
|---|---|---|---|
| Use StyleNeRF instead of GIRAFFE | 3D-aware image generation · FFHQ 256² | FID: 35 -> 8 | Helped |
| Use StyleNeRF instead of GIRAFFE | 3D-aware image generation · AFHQ 256² | FID: 31 -> 14 | Helped |
| Use StyleNeRF instead of GIRAFFE | 3D-aware image generation · CompCars 256² | FID: 32 -> 8 | Helped |
| Remove progressive training (ablation) | StyleNeRF · 3D geometry and multi-view synthesis | 3D consistency | Hurt |
| Remove NeRF-path regularization (ablation) | StyleNeRF · 3D geometry and multi-view synthesis | 3D consistency | Hurt |
| Add view direction as a color-prediction condition (ablation) | StyleNeRF · Multi-view synthesis | 3D consistency | Hurt |
| Replace the proposed upsampler with pixel-wise learnable upsampling (ablation) | StyleNeRF · Multi-view image synthesis | 3D consistency | Hurt |
| Replace the proposed upsampler with fixed bilinear interpolation (ablation) | StyleNeRF · Image synthesis | visual quality | Hurt |
| Remove per-pixel noise injection (ablation) | StyleNeRF · Image synthesis | FID | Hurt |
| Use geometry-aware noise injection instead of no noise injection (ablation) | StyleNeRF at inference · Image synthesis | stochastic detail | Helped |
| Train using progressive growing rather than progressive up-sampling (ablation) | StyleNeRF · Low-resolution training | rendering speed | Helped |
Closest papers by meaning
- StyleMorph: Disentangled 3D-Aware Image Synthesis with a 3D Morphable StyleGAN (ICLR 2023 · similarity 0.90)
- StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generation (CVPR 2022 · similarity 0.90)
- FPRF: Feed-Forward Photorealistic Style Transfer of Large-Scale 3D Neural Radiance Fields (AAAI 2024 · similarity 0.88)
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.87)
- Dual Mapping of 2D StyleGAN for 3D-Aware Image Generation and Manipulation (Student Abstract) (AAAI 2024 · similarity 0.87)
- 3DToonify: Creating Your High-Fidelity 3D Stylized Avatar Easily from 2D Portrait Images (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.