Research map / 3D vision / Gaussian splatting
BEA-GS: BEyond RAdiance Supervision in 3DGS for Precise Object Extraction
Alessio Mazzucchelli, Maria Naranjo-Almeida, Jorge Bustos-Sanchez, Mariella Dimiccoli, Francesc Moreno-Noguer, Jordi Sanchez-Riera, Adrian Penate-Sanchez
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 occupancy loss to reprojection plus boundary loss (ablation) | BEA-GS · LeRF | IoU: 78.2 -> 89.4 | Improved |
| add occupancy loss to reprojection plus boundary loss (ablation) | BEA-GS · Mip-NeRF 360 | IoU: 90.5 -> 92.0 | Improved |
| add boundary loss to reprojection plus occupancy loss (ablation) | BEA-GS · LeRF | IoU: 85.0 -> 89.4 | Improved |
| add boundary loss to reprojection plus occupancy loss (ablation) | BEA-GS · Mip-NeRF 360 | BIoU: 81.9 -> 85.8 | Improved |
| add multiview reprojection to boundary plus occupancy losses (ablation) | BEA-GS · LeRF | IoU: 86.6 -> 89.4 | Improved |
| add multiview reprojection to boundary plus occupancy losses (ablation) | BEA-GS · Mip-NeRF 360 | IoU: 90.9 -> 92.0 | Improved |
| add multiview reprojection to boundary plus occupancy losses (ablation) | BEA-GS · LLFF | IoU: 93.0 -> 93.0 | No clear effect |
| add multiview reprojection to boundary plus occupancy losses (ablation) | BEA-GS · 3DOVS | IoU: 93.3 -> 93.2 | No clear effect |
| add all three components to the vanilla assignment baseline (ablation) | BEA-GS · LeRF | IoU: 66.3 -> 89.4 | Improved |
| train on noisy depth and normals (no regularization) (ablation) | BEA-GS · 3DOVS | IoU: 93.2 -> 89.8 | Hurt |
| train on noisy depth and normals (no regularization) (ablation) | BEA-GS · Mip-NeRF 360 | IoU: 92.0 -> 92.3 | No clear effect |
| increase voxel-density points k from 250 to 5000 (ablation) | BEA-GS · LeRF | IoU: 90.4 -> 88.3 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- Flashsplat — FlashSplat: 2D to 3D Gaussian Splatting Segmentation Solved Optimally (ECCV 2024) · better on 6 of 6 shared results
- ObjectGS — ObjectGS: Object-aware Scene Reconstruction and Scene Understanding via Gaussian Splatting (ICCV 2025) · better on 0 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
- Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields (CVPR 2024 · similarity 0.89)
- Textured Gaussians for Enhanced 3D Scene Appearance Modeling (CVPR 2025 · similarity 0.89)
- Bootstraping Clustering of Gaussians for View-consistent 3D Scene Understanding (AAAI 2025 · similarity 0.89)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 2025 · similarity 0.88)
- FastLGS: Speeding Up Language Embedded Gaussians with Feature Grid Mapping (AAAI 2025 · similarity 0.88)
- SceneSplat++: A Large Dataset and Comprehensive Benchmark for Language Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
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.
