Reproduce “Off the Planckian Locus: Using 2D Chromaticity to Improve In-Camera Color” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

SaiKiran Tedla, Joshua E. Little, Hakki Can Karaimer, Michael S. Brown

Researchers at York University and Samsung Electronics developed a method to improve in-camera color accuracy by using 2D chromaticity coordinates instead of 1D Correlated Color Temperature (CCT) for illuminant characterization. Their lightweight neural network, CCM-MLP, achieved an average 22% reduction in angular reproduction error on LED-lit scenes, enhancing color fidelity for modern lighting environments.

Key insights

  • Utilizing 2D chromaticity coordinates for illuminant characterization fundamentally improves colorimetric accuracy, especially for non-Planckian light sources, compared to traditional 1D CCT.
  • A computationally efficient neural network can predict robust color correction matrices for diverse lighting conditions, suitable for real-time in-camera deployment.
  • The proposed method offers superior spatial consistency and significantly lower computational cost than per-pixel direct mapping approaches, while maintaining high accuracy across various scenarios.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.