Research map / 3D vision / Gaussian splatting
FlashSplat: 2D to 3D Gaussian Splatting Segmentation Solved Optimally
Qiuhong Shen, Xingyi Yang, Xinchao Wang
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 |
|---|---|---|---|
| solve 3D-GS segmentation in closed form instead of gradient descent feature distillation | FlashSplat · LERF Figurines | extra time: 18 min -> 26 s vs SAGA; 37 min -> 26 s vs Gaussian Grouping | Improved |
| use closed-form optimal assignment rather than SAGA | FlashSplat · NVOS | mIoU (%): 90.9 -> 91.8 | Improved |
| add positive background bias gamma 0.4 (ablation) | FlashSplat · T&T truck | mIoU (%): 92.3 -> 94.2 | Improved |
| use strongly negative background bias gamma -0.8 (ablation) | FlashSplat · T&T truck | mIoU (%): 92.3 -> 82.4 | Hurt |
| reduce number of 2D masks to 1/8 of views (ablation) | FlashSplat · 360-degree scenes | qualitative segmentation | No clear effect |
| reduce number of 2D masks to 1/16 or 1/32 of views (ablation) | FlashSplat · 360-degree scenes | qualitative segmentation | Hurt |
| add alpha-mask quantization with threshold for novel view mask rendering (ablation) | FlashSplat · novel-view masks | qualitative mask quality | Improved |
| use FlashSplat instead of SAGA for peak memory | FlashSplat · LERF Figurines | peak memory: 15G -> 8G | Improved |
Later papers that compared with this method
3 later accepted papers report Flashsplat as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- BEA-GS: BEyond RAdiance Supervision in 3DGS for Precise Object Extraction (CVPR 2026 · better on 6 of 6 shared results)
- Trace3D: Consistent Segmentation Lifting via Gaussian Instance Tracing (ICCV 2025 · better on 3 of 4 shared results)
- WildSeg3D: Segment Any 3D Objects in the Wild from 2D Images (ICCV 2025 · better on 2 of 2 shared results)
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.
- SAGA — SAGA: Stochastic Whole-Body Grasping with Contact (ECCV 2022) · better on 3 of 4 shared results
- Gaussian Grouping — Gaussian Grouping: Segment and Edit Anything in 3D Scenes (ECCV 2024) · better on 2 of 2 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
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.90)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering (CVPR 2024 · similarity 0.90)
- GS-2DGS: Geometrically Supervised 2DGS for Reflective Object Reconstruction (CVPR 2025 · similarity 0.89)
- Splatter Image: Ultra-Fast Single-View 3D Reconstruction (CVPR 2024 · similarity 0.89)
- SceneSplat++: A Large Dataset and Comprehensive Benchmark for Language Gaussian Splatting (NeurIPS 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.
