Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the weighted variational model instead of the compared methods for illumination adjustment | Image enhancement · Figure 4 average | NIQE: 2.75 vs 2.79; 3.35; 2.87 | Helped |
| Use the weighted variational model instead of the compared methods for backlit images | Image enhancement · Backlit images | NIQE: 3.51 vs 3.68; 3.43; 3.37 | Mixed |
| Use the weighted variational model instead of the compared methods for non-uniform illumination images | Image enhancement · Non-uniform illumination images | NIQE: 3.17 vs 3.28; 3.25; 3.33 | Helped |
| Use the weighted variational model instead of the compared methods for night-time images | Image enhancement · Night-time images | NIQE: 4.21 vs 5.01; 4.55; 4.51 | Helped |
| Estimate reflectance and illumination simultaneously with the weighted model rather than using Kimmel's or Ng's model | Reflectance and illumination estimation · No-reference-image qualitative comparisons | reflectance detail preservation | Helped |
| Use the proposed reflectance and illumination model rather than Kimmel's or Ng's model before Gamma correction | Dark-image enhancement with additive white Gaussian noise · Figures 5 and 6 | noise suppression | Helped |
| Increase c1 from its default value (ablation) | Weighted variational model · Reflectance estimation | detail preservation | Hurt |
| Increase c2 (ablation) | Weighted variational model · Illumination estimation | illumination smoothness | Helped |
| Use a smaller stopping tolerance ε1 = ε2 (ablation) | Weighted variational model · 512 × 512 images | computational time: 0.41 -> 1.79 -> 13.51 | Hurt |
Closest papers by meaning
- Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance (CVPR 2024 · similarity 0.85)
- Illumination Spectrum Estimation for Multispectral Images via Surface Reflectance Modeling and Spatial-Spectral Feature Generation (CVPR 2025 · similarity 0.84)
- Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · similarity 0.83)
- Zero-Reference Low-Light Enhancement via Physical Quadruple Priors (CVPR 2024 · similarity 0.83)
- Generative Multiview Relighting for 3D Reconstruction under Extreme Illumination Variation (CVPR 2025 · similarity 0.82)
- Attentive Illumination Decomposition Model for Multi-Illuminant White Balancing (CVPR 2024 · similarity 0.82)
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.