This is the repo for the "Using deep learning to non-iteratively conduct palette generation in class specific color quantization" project.

Color quantization (CQ) is the process of reducing the amount of colors in an input image without losing visual clarity. This is traditionally performed using a clustering algorithm such as K-Means which have several implementation difficulties. We seek to bypass those difficulties by instead using CNNs to pick the "palette" (few colors that will be used to represent the image) given an input image.