Grand Challenge on Detecting Cheapfakes

SimulaMet's HOST department is co-organising a grand challenge on the detection of "cheapfakes" together with YouTube, the University of Bergen (UiB), and the Technical University of Munich (TUM).

SimulaMet, YouTube, UiB and TUM have announced a Grand Challenge on Detecting Cheapfakes as part of the ACM Multimedia Systems Conference (MMSys'21). The objective of the challenge is to benchmark algorithms that can be used to detect out-of-context (image, caption) pairs, or in other words, the misuse of photographs with conflicting image captions. The participants will be using a recently compiled dataset with news items.

What are "Cheapfakes"?

Cheapfake is a general term that encompases many non-AI "cheap" manipulations of multimedia content, including the use of editing software such as Photoshop, but also the alteration of context, etc. Cheapfakes are in fact known to be more prevalent than deep fakes (see here and here).

Participation in the Challenge

The challenge is open to everyone, and a great opportunity especially for PhD and master students who are interested in joining research efforts combatting multimedia misinformation.

We encourage everyone to join the challenge.

Deadlines: Register by June 25th. Upload by June 28th.
Awards: The winner will be awarded 5,000 USD and the runner-up will be awarded 2,500 USD.

More information is found on the ACM MMSys 2021 webpage.