Oklch+: New color space matches CIEDE2000 with only 3 parameters

Oklab and its cylindrical variant Oklch are popular for perceptually motivated color interpolation, but their color difference prediction accuracy lags behind the CIEDE2000 standard. Naoyuki Uchida's new paper introduces Oklch+, a minimal extension using just three parameters: a power transformation on the L (lightness) axis and a Naka-Rushton compression on the C (chroma) axis. Euclidean distance in this transformed Oklab space achieves STRESS=29.09 on the COMBVD dataset (3,813 suprathreshold color pairs), essentially matching CIEDE2000's 29.13—but with only 3 parameters instead of approximately 17. The Naka-Rushton function, bounded in [0,1], models the saturating nature of chroma sensitivity at high colorimetric values.

Cross-validation on a held-out BFD-P D65 subset (2,028 pairs) confirms generalization: Oklch+ scores STRESS=26.14, substantially outperforming Oklab's 51.45 and approaching CIEDE2000's 24.12. Improvement over Oklab is consistent across all six COMBVD sub-datasets. Because Oklch+ defines a coordinate system where Euclidean distance approximates perceptual distance, linear interpolation in the transformed space offers substantially improved perceptual uniformity—critical for gradients and color transitions in design tools. Current evaluation is limited to the sRGB-centered COMBVD dataset; validation for high-chroma regions with empirical observer data remains future work. The paper is submitted to *Color Research & Application*.