| CPC G06T 5/00 (2013.01) [G06F 18/214 (2023.01); G06V 10/56 (2022.01); G06V 10/60 (2022.01); G06T 2207/10024 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] | 19 Claims |

|
1. An image processing method for correcting colors in an input image representing a scene, said image processing method comprising:
processing the input image with a machine learning model, wherein said machine learning model is previously trained to detect a predefined number N>1 of sources of light illuminating the scene and to generate N estimated illuminant images associated respectively to the N sources of light, wherein each estimated illuminant image comprises an estimated color of the light emitted by the respective source of light and an estimated contribution image comprising pixels, wherein each pixel value of said estimated contribution image corresponds to an estimated contribution of the light emitted by the respective source of light to the total illumination, by the N sources of light, of a respective portion of the scene represented by said each pixel of the estimated contribution image;
generating a total illuminant image by using the N estimated illuminant images, wherein the total illuminant image comprises pixels, wherein each pixel value of the total illuminant image corresponds to a local color determined by combining the estimated colors of the N sources of light according to the estimated contributions of the corresponding pixels in the estimated contribution images of the N estimated illuminant images; and
generating an output image by correcting the colors in the input image based on the total illuminant image.
|