Walk through recent advances made possible by machine learning and outstanding problems the field is still dealing with.
Doctored images have been with us for as long as people have taken photographs, but the threat of image tampering – and the quality of forgeries – has never been higher. Machine learning can be both friend and foe in this fight; some techniques enable the creation of increasingly convincing forgeries, while others help us quickly and automatically detect tampered images.
In this webinar, Faculty Data Scientist Giles Shaw will discuss the algorithms and techniques that we believe are most effective for detecting image tampering, including:
- Ideas from image forensics and ‘traditional’ approaches to detecting image tampering before the advent of deep learning.
- Recent advances in the field made possible by deep learning.
- Remaining pitfalls and challenges in the field.
For any questions please email firstname.lastname@example.org
No prior knowledge of image forensics will be assumed.
date & time
Tuesday 4 August 2020