The purpose of this research software is to create an algorithm to digitally sign video in a manner that would protect video from editing in these domains:
- signal
- shape
- time
This is a timely and important research matter to avoid the spread of digital misinformation and the rise of AI generated, photorealistic videos--a true bane on a functioning democracy.
Code derived on top of NVIDIA Sample: imageDenoising Minimum spec: SM 2.0