Research map / 3D vision / Gaussian splatting
Learning 3D Geometry and Feature Consistent Gaussian Splatting for Object Removal
Yuxin Wang, Qianyi Wu, Guofeng Zhang, Dan Xu
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 cross-attention feature regularization between Gaussians in in-painted and visible regions (ablation) | GScream w/o Cross-Attn · SPIn-NeRF | masked PSNR: 15.63 -> 15.84 | Improved |
| add monocular depth supervision with online alignment (ablation) | GScream w/o Cross-Attn & Mono-Depth · SPIn-NeRF | masked PSNR: 14.87 -> 15.63 | Improved |
| add cross-attention regularization (ablation) | GScream · SPIn-NeRF | LPIPS: 0.26 -> 0.28 (masked LPIPS 0.50 -> 0.54) | Hurt |
| use 3D Gaussian Splatting (Scaffold-GS) instead of NeRF for object removal | GScream · SPIn-NeRF | training time: ~3.0h / ~6.0h -> ~1.2h | Improved |
| GScream vs View-Sub on masked LPIPS | GScream · SPIn-NeRF | masked LPIPS: 0.54 vs 0.45* | No clear effect |
| GScream vs NeRF baselines on SSIM and FID | GScream · SPIn-NeRF | SSIM, FID: SSIM 0.54 -> 0.58; FID 38.69 -> 36.72 (vs OR-NeRF) | Improved |
| swap 2D in-painting model between LaMa and Stable Diffusion for the reference view (ablation) | GScream · Qualitative | visual quality | No clear effect |
| use Marigold instead of MiDaS for monocular depth prior (ablation) | GScream · Qualitative | depth continuity | Improved |
Closest papers by meaning
- Textured Gaussians for Enhanced 3D Scene Appearance Modeling (CVPR 2025 · similarity 0.92)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.91)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.91)
- VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment (NeurIPS 2025 · similarity 0.91)
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images (CVPR 2025 · 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.
