Research map / 3D vision / NeRF and radiance fields
SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis
Guangcong Wang, Zhaoxi Chen, Chen Change Loy, Ziwei Liu
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 local depth ranking distillation (ablation) | RegNeRF baseline · LLFF (3 views) | PSNR: 19.08 -> 19.86 | Improved |
| use depth ranking instead of scale-invariant depth scaling (ablation) | SparseNeRF · LLFF (3 views) | PSNR: 18.45 -> 19.86 | Improved |
| add depth scaling to baseline (ablation) | RegNeRF baseline · LLFF (3 views) | PSNR: 19.08 -> 18.45 | Hurt |
| use DPT Hybrid or Large instead of MiDaS small depth model (ablation) | SparseNeRF · LLFF (3 views) | PSNR: 19.35 -> 19.86 | Improved |
| use DPT Large instead of DPT Hybrid (ablation) | SparseNeRF · LLFF (3 views) | PSNR: 19.86 -> 19.86 | No clear effect |
| add depth distillation on top of FreeNeRF | FreeNeRF · LLFF (3 views) | PSNR: 19.60 -> 20.26 | Improved |
| add spatial continuity regularization (ablation) | SparseNeRF · LLFF user study | user preference: 88.9% prefer w/ continuity vs 11.1% | Improved |
| use depth-distilled continuity term instead of RegNeRF's smoothness term (ablation) | baseline · LLFF user study | user preference: 87.5% vs 12.5% | Improved |
| replace RegNeRF with SparseNeRF on DTU | SparseNeRF · DTU (3 views) | LPIPS: 0.190 -> 0.201 | Mixed |
Later papers that compared with this method
12 later accepted papers report SparseNeRF as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.
- Empowering Sparse-Input Neural Radiance Fields with Dual-Level Semantic Guidance from Dense Novel Views (AAAI 2026 · better on 1 of 2 shared results)
- RDG-GS: Relative Depth Guidance with Gaussian Splatting for Real-time Sparse-View 3D Rendering (IJCV 2026 · better on 1 of 1 shared result)
- FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors (CVPR 2025 · better on 5 of 7 shared results)
- NVS-Solver: Video Diffusion Model as Zero-Shot Novel View Synthesizer (ICLR 2025 · better on 2 of 2 shared results)
- Sparse-DeRF: Deblurred Neural Radiance Fields From Sparse View (TPAMI 2025 · better on 0 of 1 shared result)
- CoherentGS: Sparse Novel View Synthesis with Coherent 3D Gaussians (ECCV 2024 · better on 2 of 5 shared results)
- MVPGS: Excavating Multi-view Priors for Gaussian Splatting from Sparse Input Views (ECCV 2024 · better on 4 of 4 shared results)
- DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization (CVPR 2024 · better on 0 of 3 shared results)
4 more not listed.
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.
- FreeNeRF — FreeNeRF: Improving Few-Shot Neural Rendering With Free Frequency Regularization (CVPR 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
- DäRF: Boosting Radiance Fields from Sparse Input Views with Monocular Depth Adaptation (NeurIPS 2023 · similarity 0.93)
- AltNeRF: Learning Robust Neural Radiance Field via Alternating Depth-Pose Optimization (AAAI 2024 · similarity 0.92)
- ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 2024 · similarity 0.90)
- FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors (CVPR 2025 · similarity 0.89)
- Global and Hierarchical Geometry Consistency Priors for Few-shot NeRFs in Indoor Scenes (CVPR 2024 · similarity 0.89)
- ZeroRF: Fast Sparse View 360deg Reconstruction with Zero Pretraining (CVPR 2024 · 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.
