Research map / 3D vision / Gaussian splatting
CoR-GS: Sparse-View 3D Gaussian Splatting via Co-Regularization
Jiawei Zhang, Jiahe Li, Xiaohan Yu, Lei Huang, Lin Gu, Jin Zheng, Xiao Bai
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 co-pruning (prune Gaussians that do not match between two fields) (ablation) | 3DGS · LLFF (3 views) | PSNR: 19.22 -> 19.62 | Improved |
| add pseudo-view co-regularization (ablation) | 3DGS · LLFF (3 views) | PSNR: 19.22 -> 20.26 | Improved |
| combine co-pruning with pseudo-view co-regularization (ablation) | CoR-GS · LLFF (3 views) | PSNR: 20.26 -> 20.45 | Improved |
| add both co-pruning and pseudo-view co-regularization (ablation) | 3DGS · DTU (3 views) | PSNR: 17.65 -> 19.21 | Improved |
| apply co-regularization on 360-degree scenes with 12 views | CoR-GS · Mip-NeRF360 (12 views) | LPIPS: 0.415 -> 0.418 | Mixed |
| apply CoR-GS with only 3 DTU views | CoR-GS · DTU (3 views) | PSNR: 19.21 vs FreeNeRF 19.92 | Mixed |
| add co-regularization (two fields, pseudo views) to 3DGS | 3DGS · LLFF (3 views) efficiency | number of Gaussians / FPS: 1.16x10^5 -> 7.85x10^4 points; 318 -> 349 FPS | Improved |
| train two Gaussian fields and render pseudo views | 3DGS · LLFF (3 views) efficiency | training time: 2.5min -> 6min | Hurt |
| co-pruning distance threshold tau=3 instead of 5 (ablation) | CoR-GS · LLFF (3 views) | novel view quality | Hurt |
Later papers that compared with this method
23 later accepted papers report CoR-GS as a baseline in their results tables; 23 of them report a better number on most of the results they share with it.
- D$^2$GS: Depth-and-Density Guided Gaussian Splatting for Stable and Accurate Sparse-View Reconstruction (ICLR 2026 · better on 5 of 5 shared results)
- TWINGS: Thin Plate Splines Warp-aligned Initialization for Sparse-View Gaussian Splatting (CVPR 2026 · better on 5 of 5 shared results)
- Sparse4DGS: 4D Gaussian Splatting for Sparse-Frame Dynamic Scene Reconstruction (AAAI 2026 · better on 5 of 5 shared results)
- Reconstruction Using the Invisible: Intuition from NIR and Metadata for Enhanced 3D Gaussian Splatting (AAAI 2026 · better on 4 of 4 shared results)
- Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery (ECCV 2026 · better on 4 of 4 shared results)
- RDG-GS: Relative Depth Guidance with Gaussian Splatting for Real-time Sparse-View 3D Rendering (IJCV 2026 · better on 4 of 4 shared results)
- HeroGS: Hierarchical Guidance for Robust 3D Gaussian Splatting under Sparse Views (CVPR 2026 · better on 3 of 3 shared results)
- SparseSurf: Sparse-View 3D Gaussian Splatting for Surface Reconstruction (AAAI 2026 · better on 3 of 3 shared results)
15 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- FreeNeRF — FreeNeRF: Improving Few-Shot Neural Rendering With Free Frequency Regularization (CVPR 2023) · better on 0 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
- Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.95)
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.92)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.91)
- PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting (CVPR 2025 · similarity 0.91)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
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.
