In this work, we propose a biologically plausible computational framework for color perception.

What is it about?

Our biologically plausible computational framework for color perception initiates with simulating the responses of single and double opponent cells to a visual stimulus in chromatic and achromatic channels. The double opponent and the intensity channels are reconstructed using spiking neural networks and linearly combined with the single opponent channels to provide the perceived image.

Featured Image

Photo by Robert Katzki on Unsplash

Photo by Robert Katzki on Unsplash

Why is it important?

Our model allows the attribution of perceptual differences to the proportions between the single and double opponent cells’ activity, while being general enough to account for a wide range of visual phenomena including color constancy, color assimilation, and ambiguous color perception.