Augmenting colour communication in English, Greek and Thai
Mylonas, Dimitris, Koliousis, Alexandros, Stutters, Jonathan, Katemake, Pichayada, Stockman, Andrew and Eskew, Rhea T. (2023) Augmenting colour communication in English, Greek and Thai. In: 15th AIC Congress, 28 November to 02 December 2023, Chinag Rai, Thailand.
Abstract
The number of colour names varies across languages. To augment colour communication between speakers of different languages, we need a multilingual method to map how we perceive colours to the words we use to describe them. We evaluate the performance of a supervised colour naming model, Rotated Split Trees (RST), trained by responses from a crowdsourced colour naming experiment in English, Greek and Thai. We assess the generalizability of the model across several colour spaces where it performed best in the CIELUV. A comparison of RST with previous computational colour naming methods using independent psychophysical data in English showed that RST achieves state-of-the-art performance for basic colour categories and identifies five additional categories (n = 16) on the surface of the Munsell system. A demonstration of the performance of RST in segmenting a synthetic image across the colour gamut into colour names in English (n = 30), Greek (n = 28) and Thai (n = 46) further supports earlier findings that speakers can identify 30-50 colour names in their native language.
Actions (login required)
 |
Edit Item |