Title: Computer-vision-informed security and forensics

Abstract:
In this session, we will discuss how computer vision has impacted the fields of security and forensics. Specifically, we will discuss the current challenges in detecting synthetic videos and images generated by the most recent high-resolution generative AI models, including the Text-To-Video (T2V) and Image-To-Video (I2V) models. We will present a new dataset aimed at promoting the development of robust and generalizable detection systems for T2V and I2V videos. Our discussions will also include how synthetic face images can be used in security applications where the goal is to predict the appearance of an individual across time.