Research map / 3D vision / NeRF and radiance fields
MixNeRF: Modeling a Ray With Mixture Density for Novel View Synthesis From Sparse Inputs
Seunghyeon Seo, Donghoon Han, Yeonjin Chang, Nojun Kwak
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 |
|---|---|---|---|
| Model colors along ray samples with a mixture density model (ablation) | MixNeRF · Sparse-view novel view synthesis | rendering quality | Improved |
| Add auxiliary ray-depth estimation (ablation) | MixNeRF with mixture density modeling · Sparse-view novel view synthesis | rendering quality | Improved |
| Add regenerated blending weights without ray-depth supervision (ablation) | MixNeRF · Sparse-view novel view synthesis | rendering quality | No clear effect |
| Add regenerated blending weights based on estimated ray depth (ablation) | MixNeRF with ray-depth supervision · Sparse-view novel view synthesis | rendering quality | Improved |
| Reduce ray samples from 128 to 32 (ablation) | MixNeRF · LLFF 3-view | SSIM: 0.629 -> 0.620 | Hurt |
| Use 32 rather than 128 ray samples | MixNeRF versus RegNeRF · LLFF 3-view | SSIM: 0.576 -> 0.620 | Improved |
| Use 16 rather than 128 ray samples (ablation) | MixNeRF · LLFF 3-view | SSIM: 0.629 -> 0.580 | Hurt |
| Use mixture-density modeling instead of RegNeRF's approach | MixNeRF versus RegNeRF · LLFF 3-view | PSNR: 19.08 -> 19.27 | Improved |
| Use mixture-density modeling instead of RegNeRF's approach | MixNeRF versus RegNeRF · Realistic Synthetic 360° 4-view | PSNR: 13.71 -> 18.99 | Improved |
Later papers that compared with this method
3 later accepted papers report MixNeRF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- MCGS: Multiview Consistency Enhancement for Sparse-View 3D Gaussian Radiance Fields (TPAMI 2025 · better on 1 of 1 shared result)
- Few-shot NeRF by Adaptive Rendering Loss Regularization (ECCV 2024 · better on 1 of 1 shared result)
- FlipNeRF: Flipped Reflection Rays for Few-shot Novel View Synthesis (ICCV 2023 · better on 7 of 7 shared results)
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.
- InfoNeRF — InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Rendering (CVPR 2022) · better on 3 of 3 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
- RegNeRF: Regularizing Neural Radiance Fields for View Synthesis From Sparse Inputs (CVPR 2022 · similarity 0.90)
- Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenes (CVPR 2021 · similarity 0.90)
- ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 2024 · similarity 0.89)
- Few-Shot Neural Radiance Fields under Unconstrained Illumination (AAAI 2024 · similarity 0.87)
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.87)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.87)
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.
