Volume was a machine learning tool for storytellers I worked in collaboration with Or Fleisher.

Using state-of-the-art machine learning research, Volume was able to reconstruct a single 2D image or video in 3D space. Volume was built with hope to become an accessible solution, to allow anyone to easily generate a 3D asset and use it in 3D environments- to be viewed in virtual, augmented and mixed reality platforms.


Our most notable experiment was “inside pulp fiction,” which went viral within hours. In this experiment we used Volume’s teechnology to reconstrut Pulp Fiction’s iconic dance scene in Agmented Reality. 



︎ Launch GitHub 

Date: July 2017- September 2018Developed together with Or Fleisher
Technology: TensorFlow, Volume.
Featured:  The Next Web, Daily Planet Discovery Channel, Tecmundo, SIGGRAPH 2018, The Next Web (2), Vice, Prostethic Knowledge, Mashable, UploadVR, VRFocuse, Android Headlines, Labroots, Arporn, UnMediated Journal of Politics and Communication.

︎   ︎