# SAFE: Scene-Aware Feature Modulation for Color Constancy with Learned Color Space in Pure-Color Scenes

arXiv: https://arxiv.org/abs/2608.13967 · HTML: https://mlsanity.com/paper/2608.13967/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 31 of its 65 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Builds on (most cited first, 30 of 31)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform](https://mlsanity.com/paper/1804.02815/) (2018)
- [Spatially-Adaptive Feature Modulation for Efficient Image Super-Resolution](https://mlsanity.com/paper/2302.13800/) (2023)
- [Deep White-Balance Editing](https://mlsanity.com/paper/2004.01354/) (2020)
- [Realistic Blur Synthesis for Learning Image Deblurring](https://mlsanity.com/paper/2202.08771/) (2022)
- [Fast Fourier Color Constancy](https://mlsanity.com/paper/1611.07596/) (2016)
- [Multi-stage image denoising with the wavelet transform](https://mlsanity.com/paper/2209.12394/) (2022)
- [Convolutional Color Constancy](https://mlsanity.com/paper/1507.00410/) (2015)
- [IDR: Self-Supervised Image Denoising via Iterative Data Refinement](https://mlsanity.com/paper/2111.14358/) (2021)
- [Semantically Contrastive Learning for Low-light Image Enhancement](https://mlsanity.com/paper/2112.06451/) (2021)
- [Cross-Camera Convolutional Color Constancy](https://mlsanity.com/paper/2011.11890/) (2020)
- [MM-BSN: Self-Supervised Image Denoising for Real-World with Multi-Mask based on Blind-Spot Network](https://mlsanity.com/paper/2304.01598/) (2023)
- [CLCC: Contrastive Learning for Color Constancy](https://mlsanity.com/paper/2106.04989/) (2021)
- [On Finding Gray Pixels](https://mlsanity.com/paper/1901.03198/) (2019)
- [ParamISP: Learned Forward and Inverse ISPs using Camera Parameters](https://mlsanity.com/paper/2312.13313/) (2023)
- [Cascading Convolutional Color Constancy](https://mlsanity.com/paper/1912.11180/) (2019)
- [Auto White-Balance Correction for Mixed-Illuminant Scenes](https://mlsanity.com/paper/2109.08750/) (2021)
- [Attentive Illumination Decomposition Model for Multi-Illuminant White Balancing](https://mlsanity.com/paper/2402.18277/) (2024)
- [GCC: Generative Color Constancy via Diffusing a Color Checker](https://mlsanity.com/paper/2502.17435/) (2025)
- [CCMNet: Leveraging Calibrated Color Correction Matrices for Cross-Camera Color Constancy](https://mlsanity.com/paper/2504.07959/) (2025)
- [Revisiting Image Fusion for Multi-Illuminant White-Balance Correction](https://mlsanity.com/paper/2503.14774/) (2025)
- [Pixel-Wise Color Constancy via Smoothness Techniques in Multi-Illuminant Scenes](https://mlsanity.com/paper/2402.02922/) (2024)
- [Integral Fast Fourier Color Constancy](https://mlsanity.com/paper/2502.03494/) (2025)
- [Efficient RAW Image Deblurring with Adaptive Frequency Modulation](https://mlsanity.com/paper/2505.24407/) (2025)
- [Perception-Inspired Color Space Design for Photo White Balance Editing](https://mlsanity.com/paper/2512.09383/) (2025)
- [RL-AWB: Deep Reinforcement Learning for Auto White Balance Correction in Low-Light Night-time Scenes](https://mlsanity.com/paper/2601.05249/) (2026)
- [A neural field model for color perception unifying assimilation and contrast](https://mlsanity.com/paper/1810.12898/) (2018)
- [Multi-illuminant Color Constancy via Multi-scale Illuminant Estimation and Fusion](https://mlsanity.com/paper/2502.02021/) (2025)

## Abstract (arXiv)

Color constancy on pure-color scenes is challenging: when most pixels share a narrow band of hues, every chromaticity-based cue collapses to a single point and standard estimators become ambiguous. We propose a compact framework that couples two innovations: (i) SAFE, a Scene-Aware FeaturE modulation network that organizes illumination cues into a structured four-token representation, which is then selectively reweighted based on scene complexity features; (ii) the Learned Color Space (LCS), a scene-dependent chromaticity normalization that directly addresses the chromaticity collapse problem for pure-color scenes. Experiment results show that SAFE consistently improves performance in pure-color scenes. Compared to the best-performing baseline in each metric, it reduces the mean angular error by 10%, the best-25% error by 20%, and the worst-25% error by 5.8%.