Research map / 3D vision / Gaussian splatting
Disco-GS: Gaussian Splatting in Dynamic Color Lighting
Ashish Kumar, A. N. Rajagopalan
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 SSIM loss L_st between J_can and observed image (ablation) | Disco-GS · Disco dataset | PSNR: 10.63 -> 21.47 | Improved |
| add color regularizer L_col (ablation) | Disco-GS · Disco dataset | PSNR: 19.37 -> 21.47 | Improved |
| add total variation loss on E(t) (ablation) | Disco-GS · Disco dataset | PSNR: 20.24 -> 21.47 | Improved |
| add brightness regularizer L_amb (ablation) | Disco-GS · Disco dataset | PSNR: 20.64 -> 21.47 | Improved |
| add brightness factor B to canonical image (output I(v) times B) (ablation) | Disco-GS · Disco dataset | PSNR: 21.03 -> 21.47 | Improved |
| decay eta from 0.95 to 0.5 instead of fixing eta=1 in SSIM loss (ablation) | Disco-GS · Disco dataset | PSNR: 18.58 -> 21.47 | Improved |
| add regularizers on top of photometric loss only (ablation) | Disco-GS · Disco dataset | PSNR: 11.76 -> 21.47 | Improved |
| replace 3DGS with Disco-GS | 3DGS · Disco dataset: Mini-library | PSNR: 14.44 -> 19.90 | Improved |
| replace 3DGS with Disco-GS | 3DGS · Disco dataset: 3D-Artwork | SSIM: 0.88 -> 0.89 | No clear effect |
| replace Wild-Gaussians with Disco-GS | Wild-Gaussians · Disco dataset: Newspaper-room | LPIPS: 0.25 -> 0.23 | Improved |
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.
- Wild-Gaussians — WildGaussians: 3D Gaussian Splatting In the Wild (NeurIPS 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
- Luminance-GS: Adapting 3D Gaussian Splatting to Challenging Lighting Conditions with View-Adaptive Curve Adjustment (CVPR 2025 · similarity 0.89)
- GI-GS: Global Illumination Decomposition on Gaussian Splatting for Inverse Rendering (ICLR 2025 · similarity 0.88)
- BARD-GS: Blur-Aware Reconstruction of Dynamic Scenes via Gaussian Splatting (CVPR 2025 · similarity 0.88)
- MetaGS: A Meta-Learned Gaussian-Phong Model for Out-of-Distribution 3D Scene Relighting (NeurIPS 2025 · similarity 0.88)
- IBGS: Image-Based Gaussian Splatting (NeurIPS 2025 · similarity 0.88)
- GS-IR: 3D Gaussian Splatting for Inverse Rendering (CVPR 2024 · 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.
