Development of Color Restoration Method Based on Deep Learning
Choi, Hyeongseok (Image Processing & Intelligent Systems Laboratory)
The recent display devices such as smart TV and PC monitor provide the color gamut of DCI-P3, which is wider than that of sRGB. However, the sRGB image displayed on the devices having a wide color gamut shows a remarkable difference with lower color saturation. To solve this problem, we extend the color gamut of sRGB to that of DCI-P3 using a convolutional neural network to present vivid color components.