Live video frames from the camera are stylized to look like famous paintings.

OpenRTiST: Real-Time Style Transfer

OpenRTiST utilizes Gabriel, a platform for wearable cognitive assistance applications, to transform the live video from a mobile client into the styles of various artworks. The frames are streamed to a server where the chosen style is applied and the transformed images are returned to the client.

Prerequisites

OpenRTiST requires a server running the backend application to connect with. The backend can run on CPU, however a machine with a discrete GPU or an integrated Intel GPU will be accelerated. Please see https://github.com/cmusatyalab/openrtist for instructions on how to setup the server.