Research map / 3D vision / NeRF and radiance fields
HDR-NeRF: High Dynamic Range Neural Radiance Fields
Xin Huang, Qi Zhang, Ying Feng, Hongdong Li, Xuan Wang, Qing Wang
Check 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 |
|---|---|---|---|
| Model the camera response with separate tone-mapper MLPs for RGB channels rather than one shared MLP (ablation) | HDR-NeRF · Real scenes, LDR-OE and LDR-NE | PSNR: 30.37 -> 31.63; 29.37 -> 31.43 | Improved |
| Add unit exposure loss to constrain the radiance scale (ablation) | HDR-NeRF · Synthetic scenes, LDR-OE | PSNR: 36.48 -> 37.52 | Improved |
| Add unit exposure loss to constrain the radiance scale (ablation) | HDR-NeRF · Synthetic scenes, LDR-NE | PSNR: 34.77 -> 35.73 | Improved |
| Add unit exposure loss to constrain the radiance scale (ablation) | HDR-NeRF · Synthetic scenes, HDR | PSNR: 13.35 -> 37.60 | Improved |
| Increase the number of input exposures from 2 to 3 (ablation) | HDR-NeRF · Synthetic scenes, LDR-OE | PSNR: 32.39 -> 37.52 | Improved |
| Increase the number of input exposures from 2 to 3 (ablation) | HDR-NeRF · Synthetic scenes, LDR-NE | PSNR: 32.76 -> 35.73 | Improved |
| Increase the number of input exposures from 2 to 3 (ablation) | HDR-NeRF · Synthetic scenes, HDR | PSNR: 33.00 -> 37.60 | Improved |
| Increase the number of input exposures from 3 to 5 (ablation) | HDR-NeRF · Synthetic scenes, LDR-OE | PSNR: 37.52 -> 37.73 | No clear effect |
| Increase the number of input exposures from 3 to 5 (ablation) | HDR-NeRF · Synthetic scenes, LDR-NE | PSNR: 35.73 -> 36.26 | Improved |
| Increase the number of input exposures from 3 to 5 (ablation) | HDR-NeRF · Synthetic scenes, HDR | PSNR: 37.60 -> 37.86 | No clear effect |
| Recover the field from multiple exposure-varied LDR views with learned tone mapping instead of standard NeRF | HDR-NeRF · Synthetic scenes, LDR-OE | PSNR: 13.97 -> 39.07 | Improved |
| Recover the field from multiple exposure-varied LDR views with learned tone mapping instead of NeRF-W | HDR-NeRF · Real scenes, LDR-OE | PSNR: 28.55 -> 31.63 | Improved |
Later papers that compared with this method
7 later accepted papers report HDR-NeRF as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.
- Expo-GS: Exposure-Aware Signed Distance Function in Gaussian Splatting for High Dynamic Range (ICML 2026 · better on 4 of 5 shared results)
- Physically Inspired Gaussian Splatting for HDR Novel View Synthesis (CVPR 2026 · better on 4 of 4 shared results)
- Seeing through Light and Darkness: Sensor-Physics Grounded Deblurring HDR NeRF from Single-Exposure Images and Events (CVPR 2026 · better on 4 of 4 shared results)
- High Dynamic Range Novel View Synthesis with Single Exposure (ICML 2025 · better on 6 of 6 shared results)
- Learning Neural Exposure Fields for View Synthesis (NeurIPS 2025 · better on 3 of 3 shared results)
- SEHDR: Single-Exposure HDR Novel View Synthesis via 3D Gaussian Bracketing (ICCV 2025 · better on 3 of 3 shared results)
- HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting (NeurIPS 2024 · better on 16 of 17 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
- EvHDR-NeRF: Building High Dynamic Range Radiance Fields with Single Exposure Images and Events (AAAI 2025 · similarity 0.92)
- Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images (AAAI 2024 · similarity 0.91)
- High Dynamic Range Novel View Synthesis with Single Exposure (ICML 2025 · similarity 0.91)
- Enhancing Neural Radiance Fields with Adaptive Multi-Exposure Fusion: A Bilevel Optimization Approach for Novel View Synthesis (AAAI 2024 · similarity 0.89)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.89)
- Bright-NeRF: Brightening Neural Radiance Field with Color Restoration from Low-Light RAW Images (AAAI 2025 · similarity 0.89)
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.
