Posted on 9 March 2023
However, understanding the relationship between physical properties of light and the perceived colour is a complex computational problem.
While our brain represents surface colours in a way that remains constant under different illuminations, the neural computations that lead to such a representation are unknown.
The project “From Spectra to Colour: a Big-Data Computational Approach” aims at making significant steps forward in understanding the neural mechanisms responsible for stabilising colour across different illumination conditions.
Potentially, the project could solve a long-lasting scientific debate and also lead to widespread impact on the consumer goods sector (eg white balancing algorithms for consumer digital cameras, smartphones, TV and displays, as well as AR and VR headsets), as well as improving the performance of several AI tasks, from object recognition for robotics to video surveillance.
The project, led by Dr Giuseppe Claudio Guarnera (PI) at the Department of Computer Science, in collaboration with Dr Matteo Toscani (CO-PI) at the Psychology Department, Bournemouth University, has been evaluated as a “very timely proposal”, “scientifically excellent” and “supported by an excellent interdisciplinary team”.
The project has been funded by BBSRC (total value of award £250,000), though the funding opportunity: “Supporting research in cognitive computational neuroscience”.