Research map / 3D vision / NeRF and radiance fields
Instant Uncertainty Calibration of NeRFs Using a Meta-Calibrator
Niki Amini-Naieni, Tomas Jakab, Andrea Vedaldi, Ronald Clark
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 |
|---|---|---|---|
| apply meta-calibrator to FlipNeRF uncertainties vs DANE | FlipNeRF · LLFF 3-view | calibration error: 0.0441 -> 0.0026 | Improved |
| apply meta-calibrator to FlipNeRF uncertainties vs DANE | FlipNeRF · LLFF 3-view | NLL: 3.75 -> -0.68 | Improved |
| meta-calibrator vs DANE M=10 on standard LLFF uncertainty benchmark | FlipNeRF · LLFF 4-12 views | NLL: -0.47 -> -1.67 | Improved |
| calibrate with held-out-free meta-calibrator vs holding out one view of Horns (ablation) | NeRF calibration · LLFF Horns | PSNR: 17% PSNR reduction when holding out one image | Improved |
| fit calibrator on training set instead of held-out data (ablation) | Kuleshov recalibration · LLFF | calibration error | Hurt |
| predict low-dimensional PCA coefficients instead of 384-dim discretized curve (ablation) | meta-calibrator · LLFF Fern | calibration prediction quality | Improved |
| use calibrated instead of uncalibrated uncertainties for ray selection | next-best view planning · DTU 15 scenes | PSNR | Improved |
| increase ray samples for FlipNeRF mixture up to 128 (ablation) | FlipNeRF base uncertainties · LLFF | calibration error | Improved |
| interquartile range instead of variance for uncertainty metric (ablation) | calibrated CDF · single pixel timing | time (s): 1.759 -> 0.008 | Improved |
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.
- CF-NeRF — CF-NeRF: Camera Parameter Free Neural Radiance Fields with Incremental Learning (AAAI 2024) · better on 1 of 1 shared result
- S-NeRF — S-NeRF: Neural Radiance Fields for Street Views (ICLR 2023) · 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
- Evidential Neural Radiance Fields (CVPR 2026 · similarity 0.91)
- Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields (CVPR 2024 · similarity 0.88)
- ProvNeRF: Modeling per Point Provenance in NeRFs as a Stochastic Field (NeurIPS 2024 · similarity 0.88)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.88)
- Optimize the Unseen - Fast NeRF Cleanup with Free Space Prior (NeurIPS 2025 · similarity 0.88)
- Neural Visibility Field for Uncertainty-Driven Active Mapping (CVPR 2024 · 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.
