What is it about?

We present a tool for fast, iterative generation of talking-heads from text inputs requiring only 2-3 minutes of reference video data. User can get results in under 40 seconds and can further refine the result or manipulate the performance by adding mouth expressions and changing speaking style.

Featured Image

Read the Original

This page is a summary of: Iterative Text-Based Editing of Talking-Heads Using Neural Retargeting, ACM Transactions on Graphics, August 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3449063.
You can read the full text:



The following have contributed to this page