Research map / 3D vision / NeRF and radiance fields
CodeNeRF: Disentangled Neural Radiance Fields for Object Categories
Wonbong Jang, Lourdes Agapito
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 ground-truth camera pose at test time instead of jointly optimizing pose | CodeNeRF · ShapeNet-SRN Chairs 1-view | PSNR: 22.39 -> 23.66 | Improved |
| use ground-truth camera pose at test time instead of jointly optimizing pose | CodeNeRF · ShapeNet-SRN Cars 1-view | PSNR: 22.73 -> 23.80 | Improved |
| exclude pose-estimation outliers from evaluation | CodeNeRF · ShapeNet-SRN Chairs 1-view | PSNR: 22.39 -> 23.11 | Improved |
| exclude pose-estimation outliers from evaluation | CodeNeRF · ShapeNet-SRN Cars 1-view | PSNR: 22.73 -> 23.17 | Improved |
| add a second input view | CodeNeRF (GT pose) · ShapeNet-SRN Chairs | PSNR: 23.66 -> 25.63 | Improved |
| add a second input view | CodeNeRF (GT pose) · ShapeNet-SRN Cars | PSNR: 23.80 -> 25.71 | Improved |
| CodeNeRF with GT pose versus PixelNeRF | CodeNeRF (GT pose) · ShapeNet-SRN Cars 1-view | PSNR: 23.17 -> 23.80 | Improved |
| CodeNeRF with GT pose versus PixelNeRF | CodeNeRF (GT pose) · ShapeNet-SRN Chairs 1-view | PSNR: 23.72 -> 23.66 | Hurt |
| feed viewing direction to first layer with the shape code (M1) instead of the late texture branch (ablation) | CodeNeRF · texture editing with fixed shape code | texture fidelity | Hurt |
| feed shape and texture codes jointly into the first network (single embedding, M2) instead of separate branches (ablation) | CodeNeRF · texture editing with fixed shape code | shape consistency | Hurt |
Later papers that compared with this method
5 later accepted papers report CodeNeRF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Fused-Planes: Why Train a Thousand Tri-Planes When You Can Share? (ICLR 2026 · better on 1 of 1 shared result)
- HyP-NeRF: Learning Improved NeRF Priors using a HyperNetwork (NeurIPS 2023 · better on 6 of 6 shared results)
- Novel View Synthesis with Diffusion Models (ICLR 2023 · better on 0 of 2 shared results)
- Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction (ICCV 2023 · better on 1 of 2 shared results)
- AutoRF: Learning 3D Object Radiance Fields From Single View Observations (CVPR 2022 · 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
- pixelNeRF: Neural Radiance Fields From One or Few Images (CVPR 2021 · similarity 0.90)
- CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis (AAAI 2025 · similarity 0.89)
- SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image (ECCV 2022 · similarity 0.89)
- GRF: Learning a General Radiance Field for 3D Representation and Rendering (ICCV 2021 · similarity 0.88)
- Sharf: Shape-conditioned Radiance Fields from a Single View (ICML 2021 · similarity 0.88)
- ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 2024 · 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.
