Adobe implements AI to detect Photoshopped images

Avatar By Le Williams | 2 years ago

Adobe Systems Incorporated has launched a series of new artificial intelligence and machine learning technology features, with the latest production targeted at photoshopped image detection.

Within recent weeks, the computer software company displayed AI research initiatives at the CVPR computer vision conference. Adobe illustrated how digital forensics facilitated by humans can be automated by machines in much less time.

According to Verge, an Adobe representative disclosed basic facts towards the “early-stage research project”. The company explained its role and involvement in the developing technology that helps monitor and verify the authenticity of digital media.

Adobe denotes previous and current work with law enforcement agencies, using forensics to support the search for missing children and adults as evidence of its responsible approach towards AI technology.

The new research paper explicates how machine learning can be used to distinguish three common image manipulation methods. Digital forensics experts search for clues in hidden layers of the image, exploring digital artifacts for inconsistencies and brightness.

“The benefit of these new ML approaches is that they hold the potential to discover artifacts that are not obvious and not previously known,” digital forensics expert Hany Farid told The Verge. “The drawback of these approaches is that they are only as good as the training data fed into the networks, and are, for now at least, less likely to learn higher-level artifacts like inconsistencies in the geometry of shadows and reflections.”

Adobe has more than 200 PhDs in machine learning, economics, physics, and computer science in its analytics unit. This brain power sits within Adobe’s R&D unit as well as within various departments and units.