IJCB 2023 will host three tutorial, which be held on Monday, September 25th, 2023.
Tutorial 1: Alpha and Omega of Deepfakes
Description: In this tutorial, we will discuss the evolution of deepfake manipulation – from generation to detection technology available. We will introduce the deepfake technology and its verbality to society, and create a threat to the law and order of any government. We will also highlight the recent incidents, where the broadcast of fake news has jeopardized the integrity of certain regions. Hence detecting deepfakes has earned significant attention. Consequently, it has become a hot topic of interest for governments, industry and policymakers to mitigate such vulnerabilities. Next, we will showcase the activeness of the research community from both the angle of deepfake generation and detection to resolve the threat. Finally, we will discuss the available commercial solutions and the way forward.
Tutorial 2: Face Recognition and Verification: Recent Trends and Threats
- Presenters: Guido Borghi, Nicolò Di Domenico and Lorenzo Pellegrini
- Room: Union Hall/Unionska dvorana
- Tutorial description: Tutorial Webpage
- Time: 12:00 – 14:30
Description: Face recognition technology has gained widespread prominence due to its multifaceted applications across various domains, ranging from identity verification to personalized user experiences, and therefore the need for secure and efficient facial identity verification has become crucial. Ensuring the robustness of Face Recognition and Verification systems against malicious attempts to bypass or impersonate users is vital to safeguard individuals and organizations from potential cyber threats. In particular, the Morphing Attack, i.e., the possibility to elude face verification systems through a facial morphing operation between a criminal and an accomplice, has recently emerged as a serious security threat. Despite the importance of this kind of attack, the development and comparison of Morphing Attack Detection (MAD) methods is still a challenging task, especially with solutions based on deep learning approaches. Therefore, we have developed and publicly release Revelio, a flexible and modular framework for the reproducible development and evaluation of MAD systems. After an analysis about the recent trend of Face Recognition systems available in the market and in the literature, we will show how it is possible to use the Revelio framework to develop and deploy effective Single Image MAD (S-MAD) and Differential MAD (D-MAD) algorithms on publicly released datasets, even obtaining state-of-the-art performance. Finally, we will show and illustrate the use of the FVC-onGoing and NIST MORPH FRVT, web platforms that can be used to effectively test MAD systems on sequestered datasets.
Tutorial 3: Power Papers: Some Practical Pointers
- Presenter: Terence Sim
- Tutorial Description: Tutorial webpage
- Room: White hall/Bela dvorana (main part)
- Time: 9:30 – 13:30
Description: Writing a good research paper takes effort; more so if there is a page limit. Yet this skill is required of every researcher, who, more often than not, fumbles his or her way through. Good grammar is only a start; care and craft must be applied to turn a mediocre paper into a memorable one. Writing skills can indeed be honed. From the tutor: “In this reprise talk, I will highlight the common mistakes many researchers make, and offer practical pointers to pack more punch into your paper. Needless to say, the talk will be biased: I will speak not from linguistic theories, but from personal experience, sharing what has, and has not, worked for me. I will cover the major sections of a technical paper: the Title, Introduction, Related Work, Figures and Tables, and Conclusion. I will discuss the purpose of each section, present common mistakes, and provide concrete examples of good writing. I will also show how the different sections ought to be linked to reinforce the message behind the paper. ”