Volume is a tool for reconstructing a single 2D image or video in 3D space. Using state-of-the-art machine learning research, Volume is able to generate a 3D asset from a single view. Volume is currently under development and is being built as an end-to-end solution allowing anyone to easily generate a 3D asset and use it in 3D environments. Volume is intended to encourage easy prototyping in virtual, augmented and mixed reality platforms.
Volume is my graduate thesis at the Interactive Telecommunications Program at New York University. Through my process, I developed a machine learning network, cloud API and client-side rendering software for Unity3D, Three.js, and Unreal Engine. We also launched Make Volume which is a simple interface for content creators allowing them to reconstruct 3D assets from 2D images and download packages straight into the game engine.
Developed by: Or Fleisher & Shirin Anlen