Research map / 3D vision / Gaussian splatting
GPGS: Consistent 3D Object Removal via Geometry-Aware 3D Inpainting and Projected Image Refinement in 3D Gaussian Splatting
Yongjoon Lee, Donghyeon Cho
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 scene-specific fine-tuning of Point-MAE (ablation) | GPGS geometry-aware 3D inpainting · 360-USID | FID: 151.462 -> 116.822 | Improved |
| use geometry-aware 3D inpainting instead of LaMa, Infusion or AGDD depth completion (ablation) | GPGS · 360-USID | FID: best alternative 126.758 -> 116.822 | Improved |
| add projection refinement network (ablation) | GPGS · 360-USID | FID / PSNR: FID 126.917 -> 116.822; PSNR 17.637 -> 17.972 | Improved |
| add Frequency Distribution Loss instead of L1 in refinement network (ablation) | GPGS · 360-USID | FID: 147.244 -> 116.822 | Improved |
| add intensity transfer (LAB lightness mean-std) (ablation) | GPGS · 360-USID | PSNR / SSIM: PSNR 17.813 -> 17.972; SSIM 0.436 -> 0.445 | Improved |
| GPGS instead of AuraFusion | 360 object removal · COR-NeRF | PSNR / FID: PSNR 18.832 -> 18.762; FID 76.003 -> 73.694 | Mixed |
| GPGS instead of AuraFusion | 360 object removal · 360-USID | PSNR / FID: PSNR 18.255 -> 17.972; FID 146.000 -> 116.822 | Mixed |
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.
- Gaussian Grouping — Gaussian Grouping: Segment and Edit Anything in 3D Scenes (ECCV 2024) · better on 8 of 8 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
- IndoorGS: Geometric Cues Guided Gaussian Splatting for Indoor Scene Reconstruction (CVPR 2025 · similarity 0.89)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.89)
- PlanarGS: High-Fidelity Indoor 3D Gaussian Splatting Guided by Vision-Language Planar Priors (NeurIPS 2025 · similarity 0.89)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.89)
- RestorGS: Depth-aware Gaussian Splatting for Efficient 3D Scene Restoration (CVPR 2025 · 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.
