FaceSimile: A mobile application for face image search based on interactive shape manipulation

Traditional image search methods lack efficient ways to refine search results. This creates significant problems, especially for mobile users with small screens, low bandwidth and awkward keyboard settings. In this demonstration, we present an intuitive mobile application that allows users to quickly search for face images based on geometric attributes. Users drag one or a small number of shape contours on a reference face to search for desired face images. For example, the user can drag a mouth corner to find faces with wider smiles, or the tip of the nose to find faces with a specific pose. These user-specified edits are translated by our system into a set of geometric constraints, which are used internally to quickly retrieve desired face images. Despite the powerful algorithms used in this application, we achieve real-time performance on mobile Apple devices. Our demonstration shows compelling results on a sizable database of over 10,000 faces images captured in uncontrolled environments.