Manipulated images detection

Is this viral image fake or real? Can I trust this news article with soul-catching picture of disaster, hungry child or violence? Have I seen this image before in different context? We need an approach to answer these questions!
Master

As powerful visual manipulation tools makes its away across the internet, it’s getting easier and easier to manipulate images’ content, which poses a serious problem for society. With the advent of social networking services such as Facebook, Instagram, Twitter, etc. there has been a huge increase in the volume of images generated or edited and posted to public. These images are prime sources of fake news and are often used to prove false facts and spread viral content and panic. Before action can be taken on basis of a questionable image, we must verify its authenticity. In this project, we will try to combine the best of existing image analysis tools and algorithms as well as create our own approach in order to (I) recognize manipulation on images and (II) to find the very first image was a source for one or multiple manipulations.

Goal

The goal of the thesis is to produce an algorithm and technology that can be used to detect images that was manipulated and ideally to find the original source of that manipulated image.

Learning outcome

During the course of this master thesis, the candidate will gain in-depth knowledge of Computer Vision, Machine and Deep learning, Image manipulation algorithms and Image manipulation detection methods.

Qualifications

Python or Matlab or C/C++/C# programming, Knowledge about computer vision and machine learning is an advantage

Supervisors

  • Carsten Griwodz
  • Johannes Langguth
  • Konstantin Pogorelov

References

www.fastcompany.com/90364471/adobe-has-an-ambitious-plan-to-help-the-public-spot-fake-images
theblog.adobe.com/spotting-image-manipulation-ai/