Colourify: Image Colourization using AI techniques

Converting a grayscale image into a coloured one is known as image colourization. This is a challenging task that needs human assistance to produce high-quality photos. Our model, which is an improved version of the VGG16, is what we suggest in this study. The dataset used is a subset of the MIT Places Dataset available on Kaggle. The architecture is divided into 2 parts, an automatic encoder and an automatic decoder. The model will predict the values of ‘a’ and ‘b’ in the Lab colour space. Architecture of the proposed model has been presented. The results have been compared with the existing models.