A new innovative mobile app called Mug Life leverages deep learning to let you instantly create 3D animations from any uploaded photo. “The ability to manipulate photos in Mug Life is tremendously fun and addicting,” explains Rob Cohen, Co-Founder and CEO of Mug Life. “We spent years lovingly creating Mug Life to bring people an exciting new experience and empower their creativity. I’m thrilled to see what they do with it.”
 
Users can alter the expressions of facial images by selecting a specific emoji or using pose controls to create original videos, stills or GIFs. Mug Life technology is made up of three stages: Deconstruction, Animation, and Reconstruction.
 
In the Deconstruction stage, Mug Life used a TITAN X GPU and cuDNN to train their deep learning models to analyze and decompose photos into 3D building blocks: camera, lighting, geometry, and surface texture. In the Animation stage, state-of-the-art cinema animation techniques allow users to manipulate the extracted 3D data, while preserving photorealism.
 
Finally, in the Reconstruction stage, Mug Life leverages cutting edge video game technology to re-render photos as animated 3D characters with stunning quality and detail. The entire process of transforming a photo into a 3D character is automatic, and only takes a few seconds. The app is currently only available for iOS, and Android users will need to wait until the beginning of next year.