DeepFake Manipulation Detection
CONTEXT
In the past few years, deep generative models have been coming a long way, enabling their regime to more research communities, particularly image manipulation, image enhancement, and image synthesis. Most outstandingly, deepfake techniques have made extremely excellent progress with various benefits from the development of these generative models, where the quality of facial video both in speech and visual perspectives beyond human realization. Hence, it raises a lot of serious concerns over security and privacy in media for users, experts, and even governments.
Much research has actively been carried out to propose robust algorithms that can effectively detect sophisticated deepfake manipulation from queried facial videos. Due to open access to the emergence of editing tools, such as Photoshop, we can see some rapid progress in designing more generalized deep frameworks that can adapt more deepfake styles these days.
Task
In this work, students are expected to review state-of-the-art work around this research briefly and then carefully analyze the strengths and weaknesses of respective methods. From that, they are positively encouraged to propose their own novel solutions that can significantly push the boundaries of the prior work in terms of generalization to more deepfake cases.
YOU WILL GET
- Profound experience in related fields, such as deep learning models, and actual transformation of deepfake to image region.
- A publication at top-tier AI venues if the work is qualified.
REQUIREMENTS
- Strong background in Machine Learning and Deep Learning.
- At least familiar with one of the AI frameworks, such as Tensorflow, and Pytorch.
- Having hands-on experience in basic Deep Learning projects.
- A good comprehension of related papers.
Contact:
Minh Son Nguyen, m.s.nguyen@utwente.nl