Research map / 3D vision / NeRF and radiance fields
RS-NeRF: Neural Radiance Fields from Rolling Shutter Images
Muyao Niu, Tong Chen, Yifan Zhan, Zhuoxiao Li, Xiang Ji, Yinqiang Zheng
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 trajectory smoothness regularization to vanilla RS-NeRF (ablation) | Vanilla RS-NeRF · Synthetic scenes, on-trajectory | PSNR: 21.47 -> 24.06 | Improved |
| add multi-sampling to vanilla RS-NeRF (ablation) | Vanilla RS-NeRF · Synthetic scenes, on-trajectory | PSNR: 21.47 -> 24.89 | Improved |
| add multi-sampling and smoothness regularization together (ablation) | Vanilla RS-NeRF · Synthetic scenes, out-of-trajectory | PSNR: 21.68 -> 28.65 | Improved |
| add smoothness regularization to improve camera trajectory estimate (ablation) | Vanilla RS-NeRF · Synthetic scenes | MSTE: 0.0509 -> 0.0358 | Improved |
| add multi-sampling to improve camera trajectory estimate (ablation) | Vanilla RS-NeRF · Synthetic scenes | MSTE: 0.0509 -> 0.0462 | Improved |
| use 100 poses per frame instead of 400 (ablation) | Vanilla RS-NeRF · Synthetic scenes, on-trajectory | PSNR: 24.06 -> 24.36 | Improved |
| perturb initial poses with random SE(3) noise (ablation) | RS-NeRF · Synthetic scenes, on-trajectory | PSNR: 27.93 -> 28.47 (U[-0.05, 0.05]) | No clear effect |
| use RS-NeRF versus two-stage 2D correction then NeRF | novel view synthesis from RS images · Synthetic scenes, out-of-trajectory | PSNR: 22.18 -> 28.65 (JAMNet+NeRF vs RS-NeRF) | Improved |
| use RS-NeRF versus USB-NeRF, which models per-row poses without improvements | novel view synthesis from RS images · Synthetic scenes, on-trajectory | PSNR: 25.02 -> 27.93 | Improved |
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.
- USB-NeRF — USB-NeRF: Unrolling Shutter Bundle Adjusted Neural Radiance Fields (ICLR 2024) · better on 2 of 2 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
- USB-NeRF: Unrolling Shutter Bundle Adjusted Neural Radiance Fields (ICLR 2024 · similarity 0.94)
- URS-NeRF: Unordered Rolling Shutter Bundle Adjustment for Neural Radiance Fields (ECCV 2024 · similarity 0.94)
- LuSh-NeRF: Lighting up and Sharpening NeRFs for Low-light Scenes (NeurIPS 2024 · similarity 0.92)
- UP-NeRF: Unconstrained Pose Prior-Free Neural Radiance Field (NeurIPS 2023 · similarity 0.91)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.91)
- CF-NeRF: Camera Parameter Free Neural Radiance Fields with Incremental Learning (AAAI 2024 · similarity 0.90)
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.
