Research map / 3D vision / NeRF and radiance fields
NeRFLix: High-Quality Neural View Synthesis by Learning a Degradation-Driven Inter-Viewpoint MiXer
Kun Zhou, Wenbo Li, Yi Wang, Tao Hu, Nianjuan Jiang, Xiaoguang Han, Jiangbo Lu
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 NeRFLiX to NeRF-rendered views | TensoRF, Plenoxels, NeRF-mm, NeRF, NLF, and RegNeRF · LLFF-P1 and LLFF-P2 | PSNR (dB): 26.73 -> 27.39; 26.29 -> 26.90; 22.98 -> 23.38; 26.50 -> 27.26; 27.46 -> 28.19; 19.10 -> 19.68; 23.06 -> 23.90; 24.81 -> | Improved |
| Apply NeRFLiX to rendered views | TensoRF and DIVeR · Tanks and Temples | PSNR (dB): 28.43 -> 28.94; 28.16 -> 28.61 | Improved |
| Apply NeRFLiX to rendered views | TensoRF and Plenoxels · Noisy LLFF Synthetic | PSNR (dB): 22.83 -> 24.12; 23.69 -> 25.51 | Improved |
| Apply NeRFLiX after shortened NeRF training | TensoRF and Plenoxels · LLFF-P1 | PSNR (dB): 26.18 -> 27.14; 25.73 -> 26.60 | Improved |
| Use NDS instead of BSR-simulated degradation data (ablation) | IVM · LLFF-P1 | PSNR (dB): 26.40 -> 27.39 | Improved |
| Add Re-Pos. degradation to SGN (ablation) | NDS-trained model · LLFF-P1 | PSNR (dB): 27.08 -> 27.13 | Improved |
| Add A-Blur to SGN and Re-Pos. (ablation) | NDS-trained model · LLFF-P1 | PSNR (dB): 27.13 -> 27.21 | Improved |
| Add region-adaptive degradation (ablation) | NDS-trained model · LLFF-P1 | PSNR (dB): 27.21 -> 27.39 | Improved |
| Replace random reference selection with matching-cost view selection (ablation) | IVM · LLFF | PSNR (dB): 27.06dB -> 27.39dB | Improved |
| Replace pixel-wise-only aggregation with hybrid recurrent aggregation (ablation) | IVM · LLFF-P1 | PSNR (dB): 27.13 -> 27.39 | Improved |
| Replace patch-wise-only aggregation with hybrid recurrent aggregation (ablation) | IVM · LLFF-P1 | PSNR (dB): 27.00 -> 27.39 | Improved |
| Increase hybrid aggregation from one to two iterations (ablation) | IVM · LLFF-P1 | PSNR (dB): 27.21 -> 27.33 | 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.
- Plenoxels — Plenoxels: Radiance Fields Without Neural Networks (CVPR 2022) · 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
- From NeRFLiX to NeRFLiX++: A General NeRF-Agnostic Restorer Paradigm (TPAMI 2023 · similarity 0.98)
- RefSR-NeRF: Towards High Fidelity and Super Resolution View Synthesis (CVPR 2023 · similarity 0.92)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.92)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.90)
- Optimize the Unseen - Fast NeRF Cleanup with Free Space Prior (NeurIPS 2025 · similarity 0.90)
- CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis (AAAI 2025 · 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.
