3D Character Generation from Images using Convolutional Neural Networks and 3D-Character Factory

Abstract

Face recognition using convolutional neural networks can be utilised for constructing 3D characters from a photograph or from a live person. By using convolutional neural networks human attributes like the color of the eyes, skin tone, hair color, presence or absence of facial hair, body type and so on can be identified and provided as a response to a 3D-Character in a game engine like Unity3D using factory pattern approach to make the 3D Character look like the subject.