Research map / 3D vision / Gaussian splatting
GLINT: Modeling Scene-Scale Transparency via Gaussian Radiance Transport
Youngju Na, Jaeseong Yun, Soohyun Ryu, Hyunsu Kim, Sung-Eui Yoon, Suyong Yeon
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 transmission Gaussians (Gtrans) (ablation) | GLINT · 3D-FRONT-T (synthetic) | PSNR: 32.26 -> 34.50 | Improved |
| Add reflection Gaussians (Grefl) (ablation) | GLINT · 3D-FRONT-T (synthetic) | PSNR: 32.70 -> 34.50 | Improved |
| Add transparency bootstrapping loss (Ltrans) (ablation) | GLINT · 3D-FRONT-T (synthetic) | PSNR: 33.57 -> 34.50 | Improved |
| Add normal prior loss (Lnormal) (ablation) | GLINT · 3D-FRONT-T (synthetic) | Normal MAE: 12.21 -> 7.96 | Improved |
| Add depth prior loss (Ldepth) (ablation) | GLINT · 3D-FRONT-T (synthetic) | Depth AbsRel: 0.061 -> 0.035 | Improved |
| Add geometric prior losses Lgeo (depth and normal) (ablation) | GLINT · 3D-FRONT-T (synthetic) | Normal MAE: 24.69 -> 7.96 | Improved |
| Use GLINT decomposition instead of TSGS | 3DGS-based reconstruction · 3D-FRONT-T (synthetic) | Depth AbsRel: 0.08 -> 0.04 | Improved |
| Use GLINT decomposition instead of TSGS | 3DGS-based reconstruction · DL3DV-10K | PSNR: 25.94 -> 30.21 | Improved |
| Use GLINT instead of EnvGS | 3DGS-based reconstruction · DL3DV-10K | PSNR: 29.65 -> 30.21 | Improved |
| Use GLINT instead of TSGS (normal accuracy at 22.5 deg) | 3DGS-based reconstruction · 3D-FRONT-T (synthetic) | Normal accuracy 22.5 deg: 92.24 -> 92.28 | No clear effect |
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.
- EnvGS — EnvGS: Modeling View-Dependent Appearance with Environment Gaussian (CVPR 2025) · better on 5 of 5 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
- Subsurface Scattering for Gaussian Splatting (NeurIPS 2024 · similarity 0.89)
- MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference (NeurIPS 2025 · similarity 0.89)
- Fix False Transparency by Noise Guided Splatting (NeurIPS 2025 · similarity 0.89)
- CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images (CVPR 2025 · similarity 0.88)
- Reflective Gaussian Splatting (ICLR 2025 · similarity 0.88)
- 3D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexes (CVPR 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.
