Research map / 3D vision / NeRF and radiance fields
CaesarNeRF: Calibrated Semantic Representation for Few-Shot Generalizable Neural Rendering
Haidong Zhu, Tianyu Ding, Tianyi Chen, Ilya Zharkov, Ram Nevatia, Luming Liang
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 scene-level semantic representation (length 96) to GNT (ablation) | GNT · LLFF orchid | PSNR: 20.93 -> 21.46 | Improved |
| extend GNT per-pixel feature length instead of adding semantic representation (ablation) | GNT · LLFF orchid | PSNR: 20.93 -> 20.85 | No clear effect |
| increase semantic representation length from 32 to 128 (ablation) | CaesarNeRF · LLFF orchid | PSNR: 21.43 -> 21.49 | No clear effect |
| add calibration and sequential refinement together (ablation) | CaesarNeRF with +96 semantic · LLFF orchid | PSNR: 21.46 -> 21.67 | Improved |
| add semantic calibration before averaging (ablation) | CaesarNeRF · MVImgNet (2 reference views) | PSNR: 26.34 -> 27.34 | Improved |
| uncalibrated averaging with one view (ablation) | CaesarNeRF w/o calibration · MVImgNet (1 reference view) | PSNR: 22.22 -> 23.61 | Improved |
| add Caesar pipeline to IBRNet | IBRNet · LLFF (1 reference view) | PSNR: 16.85 -> 17.76 | Improved |
| add Caesar pipeline to MatchNeRF | MatchNeRF · LLFF (2 reference views) | PSNR: 20.59 -> 21.55 | Improved |
| replace GNT with CaesarNeRF in the single-reference setting | GNT · LLFF (1 reference view) | PSNR: 16.57 -> 18.31 | Improved |
| replace GNT with CaesarNeRF with 3 reference views | GNT · mip-NeRF 360 (3 reference views) | PSNR: 15.59 -> 17.55 (IBRNet 17.70) | Mixed |
| replace GNT with CaesarNeRF in per-scene optimization | GNT · LLFF per-scene | PSNR: 27.24 -> 27.64 | Improved |
Later papers that compared with this method
1 later accepted paper reports CaesarNeRF as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis (CVPR 2025 · better on 2 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GeoNeRF — GeoNeRF: Generalizing NeRF With Geometry Priors (CVPR 2022) · 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
- GSNeRF: Generalizable Semantic Neural Radiance Fields with Enhanced 3D Scene Understanding (CVPR 2024 · similarity 0.90)
- Learning Robust Generalizable Radiance Field with Visibility and Feature Augmented Point Representation (ICLR 2024 · similarity 0.89)
- ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 2024 · similarity 0.89)
- InsertNeRF: Instilling Generalizability into NeRF with HyperNet Modules (ICLR 2024 · similarity 0.88)
- Few-Shot Neural Radiance Fields under Unconstrained Illumination (AAAI 2024 · similarity 0.88)
- FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors (CVPR 2025 · similarity 0.88)
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.
