Research map / 3D vision / Gaussian splatting
ReSplat: Degradation-agnostic Feed-forward Gaussian Splatting via Self-guided Residual Diffusion
Youngho Yoon, Kuk-Jin Yoon
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 |
|---|---|---|---|
| integrate NVS with multi-view-aware residual diffusion restoration (ReSplat vs DiffUIR then NV) | feed-forward GS (MVSGaussian) · LLFF haze+snow | PSNR: 15.38 -> 20.17 | Improved |
| ReSplat vs DiffUIR on single degradations | DiffUIR+MVSGaussian · LLFF synthetic, 5 types | PSNR: e.g. 22.75 -> 23.15 (motion blur) | Improved |
| ReSplat vs DiffUIR on low-light NVS | DiffUIR · LLFF low-light | LPIPS: 0.2429 -> 0.2433 | No clear effect |
| ReSplat vs DiffUIR on haze real-world | DiffUIR · REVIDE haze | LPIPS: 0.1900 -> 0.1968 | Mixed |
| add GS-guided multi-view alignment (ablation) | ReSplat · LLFF synthetic avg | PSNR: 22.19 -> 22.46 | Improved |
| add pre-filtering with warped features (ablation) | ReSplat · LLFF synthetic avg | PSNR: 22.19 -> 22.35 | Improved |
| combine alignment and pre-filtering (ablation) | ReSplat · LLFF synthetic avg | PSNR: 22.19 -> 22.69 | Improved |
Later papers that compared with this method
1 later accepted paper reports ReSplat as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- CodecSplat: Ultra-Compact Latent Coding for Feed-Forward 3D Gaussian Splatting (NeurIPS 2026 · better on 0 of 1 shared result)
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.
- GAURA — GAURA: Generalizable Approach for Unified Restoration and Rendering of Arbitrary Views (ECCV 2024) · better on 5 of 5 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
- RestorGS: Depth-aware Gaussian Splatting for Efficient 3D Scene Restoration (CVPR 2025 · similarity 0.90)
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.90)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.90)
- DepthSplat: Connecting Gaussian Splatting and Depth (CVPR 2025 · similarity 0.89)
- ReorgGS: Equivalent Distribution Reorganization for 3D Gaussian Splatting (NeurIPS 2026 · similarity 0.89)
- FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Training (NeurIPS 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.
