Research map / 3D vision / NeRF and radiance fields
Seeing through Light and Darkness: Sensor-Physics Grounded Deblurring HDR NeRF from Single-Exposure Images and Events
Yunshan Qi, Lin Zhu, Nan Bao, Yifan Zhao, Jia Li
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 event input supervision (ablation) | See-NeRF · Our real dataset (HDR) | PSNR: 19.84 -> 26.49 | Improved |
| add event input supervision (ablation) | See-NeRF · Our synthetic dataset (HDR) | PSNR: 22.19 -> 24.13 | Improved |
| use 2D pixel-wise RGB CRF instead of 3D points CRF (ablation) | See-NeRF · Our real dataset (HDR) | PSNR: 24.95 -> 26.49 | Improved |
| use 2D pixel-wise RGB CRF instead of 3D points CRF (ablation) | See-NeRF · Our synthetic dataset (HDR) | PSNR: 20.67 -> 24.13 | Improved |
| use MLP-based RGB CRF instead of power function CRF (ablation) | See-NeRF · Our real dataset (HDR) | PSNR: 19.75 -> 26.49 | Improved |
| add event CRF model (latency and calibration) instead of naive event model (ablation) | See-NeRF · Our real dataset (HDR) | PSNR: 25.87 -> 26.49 | Improved |
| add event CRF model instead of naive event model (ablation) | See-NeRF · Our synthetic dataset (HDR) | PSNR: 22.23 -> 24.13 | Improved |
| use our MLP event CRF instead of EvDeblur-NeRF eCRF (ablation) | See-NeRF · Our real dataset (HDR) | PSNR: 16.43 -> 26.49 | Improved |
| use our implicit MLP 2nd-order LPF instead of v2e explicit 2nd-order LPF (ablation) | See-NeRF · Real-World-Challenge | PSNR: 30.51 -> 32.70 | Improved |
| add our RGB CRF model to base without events (ablation) | See-NeRF · Real-World-Challenge | PSNR: 28.79 -> 30.22 | Improved |
| increase time samples b from 4 to 8 (ablation) | See-NeRF · Desktop HDR | PSNR: 28.13 -> 26.49 | Hurt |
| increase time samples b from 2 to 8 (ablation) | See-NeRF · Lab deblurring | PSNR: 33.98 -> 35.57 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 6 of 6 on most shared results.
- HDR-NeRF — HDR-NeRF: High Dynamic Range Neural Radiance Fields (CVPR 2022) · better on 4 of 4 shared results
- HDR-GS — HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting (NeurIPS 2024) · better on 2 of 2 shared results
- E3NeRF — E 3 NeRF: Efficient Event-Enhanced Neural Radiance Fields From Blurry Images (TPAMI 2026) · better on 1 of 1 shared result
- BAD-NeRF — BAD-NeRF: Bundle Adjusted Deblur Neural Radiance Fields (CVPR 2023) · better on 1 of 1 shared result
- DP-NeRF — DP-NeRF: Deblurred Neural Radiance Field With Physical Scene Priors (CVPR 2023) · better on 1 of 1 shared result
- E2NeRF — E2NeRF: Event Enhanced Neural Radiance Fields from Blurry Images (ICCV 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
- EvHDR-NeRF: Building High Dynamic Range Radiance Fields with Single Exposure Images and Events (AAAI 2025 · similarity 0.92)
- High Dynamic Range Novel View Synthesis with Single Exposure (ICML 2025 · similarity 0.91)
- Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images (AAAI 2024 · similarity 0.89)
- Mitigating Motion Blur in Neural Radiance Fields with Events and Frames (CVPR 2024 · similarity 0.89)
- HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting (NeurIPS 2024 · similarity 0.88)
- DynaMoDe-NeRF: Motion-aware Deblurring Neural Radiance Field for Dynamic Scenes (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.
