Research map / 3D vision / Gaussian splatting
PriSplat: Propagating Reliable Multi-view Information for Distractor-Free 3DGS
Yunseo Yang, 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 |
|---|---|---|---|
| add harmonization with geometry-aware selection and threshold (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.86 -> 23.11 | Improved |
| harmonize all views with KNN supports and no threshold (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.86 -> 22.72 | Hurt |
| use geometry-aware instead of KNN support selection (with threshold) (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.89 -> 23.11 | Improved |
| add threshold on well-supported views (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.85 -> 23.11 | Improved |
| continue training the mask predictor (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.89 -> 23.11 | Improved |
| add mask-aware fast-weight update (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.93 -> 23.11 | Improved |
| add mask-aware fast-weight update (ablation) | PriSplat · On-the-go (mean) | LPIPS: 0.1373 -> 0.1344 | Improved |
| restrict harmonization to iterations 10k-20k instead of 10k-30k (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.92 -> 23.11 | Improved |
| set Kmin=8 instead of 16 (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.95 -> 23.11 | Improved |
| combine information density, spatial and angular scores vs single score (ablation) | PriSplat · On-the-go (mean) | PSNR: 22.79-22.90 -> 23.11 | Improved |
| use PriSplat's NVS prior instead of LeftRefill inpainting prior | PriSplat · On-the-go (mean) | PSNR: 22.78 -> 23.11 | 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.
- DeSplat — DeSplat: Decomposed Gaussian Splatting for Distractor-Free Rendering (CVPR 2025) · better on 6 of 6 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
- DeSplat: Decomposed Gaussian Splatting for Distractor-Free Rendering (CVPR 2025 · similarity 0.91)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.91)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (CVPR 2025 · similarity 0.90)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis (CVPR 2025 · similarity 0.90)
- LP-3DGS: Learning to Prune 3D Gaussian Splatting (NeurIPS 2024 · 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.
