This code implementation is only for research or education purposes, and (especially the learned data) not available for commercial use, redistribution, sublicensing etc. The intellectual property and code implementation belongs to the University of Edinburgh. For scientific use, please reference this repository together with the PFNN paper below. In any case, I would ask you to contact me if you intend to seriously use, redistribute or publish anything related to this code or repository.
This project explores the opportunities of deep learning and biologically-inspired optimisation for character animation and control as part of my Ph.D. research at the University of Edinburgh in the School of Informatics, supervised by Taku Komura. The development is done using Unity3D / Tensorflow, and the implementations are made available during my Ph.D. progress.
Camera-ready version in progress.
This work continues the recent work on PFNN (Phase-Functioned Neural Networks) for character control.
Video: https://www.youtube.com/watch?v=Ul0Gilv5wvY
Paper: http://theorangeduck.com/media/uploads/other_stuff/phasefunction.pdf
A demo in Unity3D using the original weights is contained in the Assets/Demo/SIGGRAPH_2017 folder.
In progress.