Announcements
- April 30th: TWYN will be held in conjunction with ECCV 2024 - website is online!
- May 22th: CMT is online and we are officially accepting papers!
The Workshop
In an era of rapid advancements in Artificial Intelligence, the imperative to foster Trustworthy AI has never been more critical. The first “Trust What You learN (TWYN)” workshop seeks to create a dynamic forum for researchers, practitioners, and industry experts to explore and advance the intersection of Trustworthy AI and DeepFake Analysis within the realm of Computer Vision. The workshop aims to delve into the multifaceted dimensions of building AI systems that are not only technically proficient but also ethical, transparent, and accountable. The dual focus on Trustworthy AI and DeepFake Analysis reflects the workshop’s commitment to addressing the challenges posed by the proliferation of deep fake technologies while simultaneously promoting responsible AI practices.
Tracks
The workshop will consist of two primary tracks: “From Learning to Unlearning: The Role of Privacy in Computer Vision” and “DeepFake Analysis and Detection”.
Track 1 - From Learning to Unlearning: The Role of Privacy in Computer Vision
The First track invites participants to explore the intricate interplay between learning and unlearning, and the critical role of privacy in shaping the future of Computer Vision. Exploring the ethical facets of data-driven learning, we will uncover the challenges presented by intrusive vision technologies and discuss innovative strategies to unlearn biases or sensitive information deeply ingrained in these systems. The topics covered by this track are: Differential privacy; Statistical and information-theoretic notions of privacy; Privacy-preserving data sharing, anonymization, privacy of synthetic data and distillation; Privacy attacks; Federated and decentralized privacy-preserving algorithms; Privacy and bias correction in generative models; Privacy in autonomous systems; Privacy and private learning in computer vision and natural language processing tasks; Relations of privacy with fairness, transparency and adversarial robustness; Machine unlearning and data-deletion; Privacy-preserving continual learning systems.
Track 2 - DeepFake Analysis and Detection
The “DeepFake Analysis and Detection” track will provide an in-depth overview of the latest methods and progress in identifying and addressing the challenges posed by deep-fake technologies. This track focuses on the rapidly changing field of synthetic media, where participants will learn advanced techniques to distinguish real from altered visual content. Topics include: Approaches for fake image detection, relying on both low-level, hand-crafted features or learnable and semantic approaches; Partially-altered fake image detection; GAN and Diffusion-based techniques with safety reassurance for image and video synthesis and generation; Video Deepfake detection and multimodal approaches to deep-fake detection; Approaches for detecting generated text and fake news, also based on multimodal analysis; Approaches and techniques for explainable deep-fake detection; Evaluation metrics for deep-fake generation and detection systems.
📃 Read the CfP and submit your paper
Confirmed Keynote Speakers
Organized with the support of
