People preserve memories of events such as birthdays, weddings, or vacations by capturing photos, often depicting groups of people. Invariably, some individuals in the image are more important than others given the context of the event. This IP proposes understanding and quantifying this concept of importance in group photos, predicting this importance and then using it to enhance real world applications. The IP gives a way to convert this abstract, subjective concept of importance into a measurable score. The IP shows a general design methodology to use the scores to train machine learning models which can then predict importance on new images. The IP addresses how to answer two questions – Given an image, who are the most important individuals in it? Given multiple images of a person, which image depicts the person in the most important role? The IP extends this to ranking people or images based on importance and hence is immediately applicable to several applications like photo sorting, photo cropping, description generation and the likes. Experiments have backed the claims of this IP regarding the benefits of incorporating importance to enhance related applications.