Research map / 3D vision / Gaussian splatting
Robust 3D-Masked Part-level Editing in 3D Gaussian Splatting with Regularized Score Distillation Sampling
Hayeon Kim, Ji Ha Jang, Se Young Chun
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 3D-GALP masks (ablation) | RoMaP baseline · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | CLIP_dir: 0.060 -> 0.162 | Improved |
| add regularized SDS second term on top of masks (ablation) | RoMaP + Mask · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | CLIP_dir: 0.162 -> 0.205 | Improved |
| add full regularized SDS loss (ablation) | RoMaP + Mask & L1hat · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | CLIP: 0.267 -> 0.277 | Improved |
| add masks (ablation) | RoMaP baseline · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | CLIP: 0.218 -> 0.228 | Improved |
| add full regularized SDS loss (ablation) | RoMaP + Mask & L1hat · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | CLIP_dir: 0.205 -> 0.205 | No clear effect |
| raise SH order from 1 to 3 (ablation) | 3D-GALP · 3D-OVS part segmentation | mIoU: 0.4777 -> 0.5587 | Improved |
| raise SH order from 3 to 4 (ablation) | 3D-GALP · 3D-OVS part segmentation | mIoU: 0.5587 -> 0.5506 | Hurt |
| use 3D-GALP instead of LeGaussian | part segmentation · 3D-OVS part segmentation | mIoU: 0.288 -> 0.559 | Improved |
| use SLaMP scheduled latent mixing instead of RF-inversion | 2D part editing · CelebAMask-HQ | B-VQA: 0.740 -> 0.758 | Improved |
| use RoMaP instead of DGE | 3D Gaussian editing · 3D editing (IN2N and NeRF-Art scenes, 75 prompts) | B-VQA: 0.497 -> 0.723 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- GaussCtrl — GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editing (ECCV 2024) · better on 4 of 4 shared results
- LangSplat — LangSplat: 3D Language Gaussian Splatting (CVPR 2024) · 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
- 3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting (ICLR 2025 · similarity 0.87)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting (CVPR 2024 · similarity 0.86)
- SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis (CVPR 2025 · similarity 0.86)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.86)
- 3DGS-Drag: Dragging Gaussians for Intuitive Point-Based 3D Editing (ICLR 2025 · similarity 0.86)
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.
