Guillaume Sarrazin

Papers

Total Citations

H-Index

About

Guillaume Sarrazin is a researcher at the intersection of robotics, computer vision, and multi-modal perception, with a focus on enabling robots to track and interact with humans in dynamic environments. His most-cited work, "Audio-Visual Variational Fusion for Multi-Person Tracking with Robots" (2019), introduces a novel variational approach to fuse auditory and visual data, significantly improving multi-person tracking accuracy in noisy, real-world settings—a critical challenge for social robots and autonomous systems. This contribution has garnered 4 citations, reflecting its niche but growing influence in the robotics community. Sarrazin’s research addresses the fundamental problem of sensor fusion, leveraging probabilistic methods to handle uncertainty and occlusion, which is essential for safe human-robot interaction. His work is particularly notable for its application to mobile robots, where robust tracking is vital for navigation and collaboration. By pioneering audio-visual fusion techniques, Sarrazin is helping to bridge the gap between robotic perception and human-centric environments, laying groundwork for more responsive and aware autonomous systems. His achievements underscore a commitment to advancing multi-modal learning and variational inference in robotics.