Research map / 3D vision / NeRF and radiance fields
RoGUENeRF: A Robust Geometry-Consistent Universal Enhancer for NeRF
Sibi Catley-Chandar, Richard Shaw, Gregory Slabaugh, Eduardo Pérez Pellitero
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 NeRFLiX enhancer to ZipNeRF | ZipNeRF · 360v2 | LPIPS: 0.1779 -> 0.2045 | Hurt |
| add RoGUENeRF enhancer to ZipNeRF | ZipNeRF · 360v2 | PSNR: 28.90 -> 29.23 | Improved |
| add RoGUENeRF enhancer to Nerfacto | Nerfacto · 360v2 | PSNR: 26.11 -> 27.45 | Improved |
| add RoGUENeRF enhancer to NeRF | NeRF · LLFF | LPIPS: 0.2389 -> 0.1495 | Improved |
| add 3D alignment module with nearest neighbours (ablation) | RoGUENeRF · LLFF (NeRF baseline) | LPIPS: 0.2245 -> 0.2078 | Improved |
| add non-rigid refinement with optical flow network (ablation) | RoGUENeRF · LLFF (NeRF baseline) | LPIPS: 0.2078 -> 0.1799 | Improved |
| add geometry-aware attention (ablation) | RoGUENeRF · LLFF (NeRF baseline) | PSNR: 26.93 -> 27.00 | Improved |
| add pre-training on LLFF with NeRF renders (ablation) | RoGUENeRF · LLFF (NeRF baseline) | PSNR: 27.00 -> 27.67 | Improved |
| add nearest neighbours to Uformer without geometry (ablation) | Uformer · 360v2 Garden Nerfacto, 1 min fine-tune | PSNR: 25.43 -> 25.39 | No clear effect |
| use full RoGUENeRF instead of Uformer, 1 minute fine-tuning (ablation) | Nerfacto · 360v2 Garden | PSNR: 25.43 -> 26.08 | Improved |
| apply enhancer to NeRFs trained with noisy poses (large noise) | Nerfacto · 360v2 Garden, large noise | MUSIQ: 32.43 -> 51.60 | Improved |
| apply RoGUENeRF to strong ZipNeRF with no pose noise | ZipNeRF · 360v2 Garden, no noise | MUSIQ: 71.38 -> 72.06 | No clear effect |
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.
- NeuS2++ — NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstruction (ICCV 2023) · better on 1 of 1 shared result
- MipNeRF360 — Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (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
- CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis (AAAI 2025 · similarity 0.89)
- FreeNeRF: Improving Few-Shot Neural Rendering With Free Frequency Regularization (CVPR 2023 · similarity 0.88)
- From NeRFLiX to NeRFLiX++: A General NeRF-Agnostic Restorer Paradigm (TPAMI 2023 · similarity 0.88)
- FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors (CVPR 2025 · similarity 0.87)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.86)
- 3DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priors (NeurIPS 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.
