Please note: This PhD seminar will take place online.
Andreea Pocol, PhD candidate
David R. Cheriton School of Computer Science
Supervisors: Professors Lesley Istead, Craig S. Kaplan
Colour palettes have applications in product, web, and interior design; image quantization; fashion; and art and colour theory. This seminar establishes a foundational understanding of colour science, moving from the basic definitions of colour space to the importance of perceptual uniformity. The discussion evaluates the technical trade-offs between additive and subtractive models, as well as the transition from device-dependent spaces like RGB and HSL to device-independent standards such as CIELAB and CIECAM02.
Building on these foundations, the talk examines the evolution of colour difference metrics—ranging from simple Euclidean distances to advanced Delta E formulations—and their implementation in complex palette similarity models. It will analyze a spectrum of mathematical approaches, including the Single, Mean, and Minimum Color Difference Models [Pan and Westland 2018], and more sophisticated structural comparisons such as Colour-based Earth Mover’s Distance [Skaff et al. 2011], Dynamic Closest Color Warping [Kim and Choi 2021], and the Hungarian method for bipartite matching [Westland et al. 2024]. It will also feature live demonstrations of Earth Mover’s Distance and the Kuhn-Munkres algorithm.