Coloring-Black-and-White-Photos

Colorization of a black and white image is a challenging topic of ongoing research in Computer Science.Image colorization is the process of taking an input black and white image and then producing an output colorized image that represents the semantic colors and tones of the input.Previously, a lot of research work was conducted on this topic.There were different methods, but unfortunately all of them needed human annotation. Deep neural network based colorization is the first computer automated colorization which doesn’t need human interaction while colorization. We reviewed existing research work on this topic,then we implemented one of the most recent works which is proposed in a paper entitled as ‘ChromaGAN: Adversarial Picture Colorization with Semantic Class Distribution’ . This is an adversarial learning colorization approach coupled with semantic information. With the semantic clues of the image, the generative network predicts the chromaticity of that image. This model is trained via a fully supervised strategy. Later, qualitative and quantitative results show the capacity of the proposed method to colorize images in a realistic way achieving state-of-the-art results.

Pretrained weight will be available at: https://drive.google.com/file/d/1tWcmbfmmlNk7hKHECamcPKY5EteFERwc/view?usp=sharing