Vincent Rabaud

Papers

Total Citations

665

H-Index

About

Vincent Rabaud is a leading researcher in robotics and computer vision, best known for his pioneering work in visual-inertial simultaneous localization and mapping (SLAM). His most influential contribution is the development of a keyframe-based visual-inertial SLAM system that leverages nonlinear optimization, a paradigm shift from the traditional filtering-based approaches. This work, published in 2013, has garnered over 600 combined citations, cementing its status as a foundational reference in the field. By demonstrating that optimization techniques could robustly fuse visual and inertial data, Rabaud helped enable more accurate and reliable state estimation for autonomous robots and drones operating in GPS-denied environments. Beyond SLAM, he has also tackled the challenging problem of pose estimation for rigid transparent objects in cluttered scenes—a notoriously difficult task due to the unique optical properties of glass and plastic. His algorithm for recognizing and estimating the pose of overlapping transparent objects, cited over 50 times, has advanced the capabilities of robotic manipulation in everyday human environments. Rabaud’s work continues to inspire new generations of researchers in sensor fusion, 3D perception, and autonomous navigation.