There are no reviews yet. Be the first to send feedback to the community and the maintainers!
Repository Details
People get their pictures taken on Google street view or when they pass by a journalist giving a report in public places, questions concerning the privacy of people visible arises. Among those image sources exposed to the public with or without our awareness, a considerable number of them contain our identity especially the bio-metric infor- mation. To solve these issue, we implement a model which can change a persons face to look like a completely different person, thus protecting their privacy. The model is trained with combination of generative adversarial networks(GAN) and autoencoders. We ensure anonymity by synthesizing GAN generated images. The generated faces are used to de-identify subjects in images or video, while preserving non-identity-related aspects of the data and consequently enabling data utilization.
