Research map / 3D vision / Depth and stereo
Large Scale Multi-Illuminant (LSMI) Dataset for Developing White Balance Algorithm Under Mixed Illumination
Dongyoung Kim, Jinwoo Kim, Seonghyeon Nam, Dongwoo Lee, Yeonkyung Lee, Nahyup Kang, Hyong-Euk Lee, ByungIn Yoo, Jae-Joon Han, Seon Joo Kim
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 |
|---|---|---|---|
| replace patch-based learning method with pixel-level U-Net | white balance model · LSMI multi-illuminant test | mean angular error: 5.56 -> 2.35 | Improved |
| replace patch-based learning method with pixel-level HDRnet | white balance model · LSMI multi-illuminant test | mean angular error: 5.56 -> 3.13 | Improved |
| replace patch-based statistic method with pixel-level U-Net | white balance model · LSMI multi-illuminant test | PSNR: 25.4 -> 39.1 | Improved |
| replace patch-based learning method with pixel-level U-Net | white balance model · LSMI single-illuminant test | mean angular error: 4.15 -> 2.95 | Improved |
| add relighting augmentation to training (ablation) | U-Net · LSMI multi-illuminant test | mean angular error: 3.43 -> 2.35 | Improved |
| add relighting augmentation to training (ablation) | U-Net · LSMI single-illuminant test | mean angular error: 3.64 -> 2.95 | Improved |
| add relighting augmentation to training (ablation) | HDRnet · LSMI multi-illuminant test | mean angular error: 3.57 -> 3.43 | Improved |
| add relighting augmentation to training (ablation) | HDRnet · LSMI single-illuminant test | mean angular error: 3.13 -> 3.16 | No clear effect |
| use natively captured LSMI instead of flash-synthesized mixing | dataset · two-illuminant scenes | mean std of illuminant coefficient: 0.0838 -> 0.1504 | Improved |
| use natively captured LSMI instead of flash-synthesized mixing | dataset · three-illuminant scenes | mean std of illuminant coefficient: 0.1118 -> 0.1633 | Improved |
Closest papers by meaning
- Attentive Illumination Decomposition Model for Multi-Illuminant White Balancing (CVPR 2024 · similarity 0.92)
- Revisiting Image Fusion for Multi-Illuminant White-Balance Correction (ICCV 2025 · similarity 0.91)
- Two Illuminant Estimation and User Correction Preference (CVPR 2016 · similarity 0.88)
- White-Balance First, Adjust Later: Cross-Camera Color Constancy via Vision-Language Evaluation (CVPR 2026 · similarity 0.84)
- Integral Fast Fourier Color Constancy (CVPR 2025 · similarity 0.84)
- Luminance-GS: Adapting 3D Gaussian Splatting to Challenging Lighting Conditions with View-Adaptive Curve Adjustment (CVPR 2025 · similarity 0.83)
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.
