Deep Learning Avatar From Image To Virtual World

Project Summary:
Designed and implemented a Pipe-Line that takes two input images of the user and creates an animatable Avatar in the virtual world.

Work Steps:
1. Surveyed existing Directions and NN to find one that fits our needs in avatar and virtual world, Chosen ק– Construct 3D Avatar from human full body image.

  1. Investigate network code and improve the following:
  1. Improved resolution of output avatar significantly using rendering method called UV-Map instead of Vertex Colors:
    1. Investigate output format
    2. Update and expand ICON source code so it supports UV map
  2. Added support to the back image using algorithms we designed and implemented.
  3. All attempts and solutions performed while thinking about optimization and efficiency while using pytorch.
  1. Investigate and found semi-automatic way to animate avatar – Mixamo
  2. Studied and built a Unity application that includes our avatar and supports Movement and a few basic movements.
  3. During the project used and learned multiple technologies:
  4. 3D modeling and rendering basics.

Program languages and libraries: Python – Pytorch, C# – Unity