Research seminars

Bio: I am a PhD Candidate working under the supervision of Prof. Belen Masia and Prof. Diego Gutierrez at Graphics and Imaging Lab, at Universidad de Zaragoza (Spain). Previously, I studied Bachelor in Computer Engineering at Universidad de Zaragoza (Spain) and Master in Computer Science with specialization in Artificial Intelligence at University of Freiburg (Germany). My research interests lie among the fields of computer graphics, perception science, computer vision and deep learning. My PhD thesis focuses on better understanding how our brains visually perceive material appearance in order to build intuitive, perceptually-based representations of appearance for visual content creation. In my free time, I love volunteering in scientific outreach activities and raising visibility of women in STEM. Webiste here.

Abstract: Our visual perception of the world is strongly influenced by how we interpret material appearance. While humans have an innate ability to recognize materials and their properties, such as glossiness or softness, these characteristics emerge from intricate and not fully understood interactions between different factors, including not only surface reflectance, but also external conditions such as illumination, geometry, or point of view. Therefore, understanding and modeling material appearance from a perceptual perspective remains a significant scientific challenge. In computer graphics, multiple material representations exist, which are often guided by the physical interactions between light and matter. However, and despite recent advances, there is still a fundamental gap between such computational models of material appearance and perceptually meaningful, human-friendly properties, limiting our ability to interact with digital imaging tools. This thesis explores how human perception can inform the learning of latent representations that are more aligned with how we see and understand material appearance, as well as how these representations can be used for developing more intuitive and controllable material authoring tools to support user-centered creative tasks. We present the contributions of this thesis from both perspectives: finding perceptually-meaningful material representations, and developing intuitive material authoring tools.

Bio: Hi, I’m Floriane Magera, I'm from the French-speaking part of Belgium. I’m currently pursuing a PhD in Computer Vision at the University of Liège, in collaboration with EVS Broadcast Equipment, a leader in live production. My research focuses on enabling augmented reality for sports content, a task that relies heavily on precise camera calibration. So far, my work has been integrated into a VAR (Video Assistant Referee) system for football.

Abstract: Camera calibration is central to many sports technologies we now take for granted, including player tracking, match statistics, and officiating. In this talk, I’ll share insights from my industrial PhD—where academic research meets the practical challenges of real-world sports production.