Jean‐Charles Bazin

ETH Zurich, The University of Tokyo, Korea Advanced Institute of Science and Technology

Papers

Total Citations

221

H-Index

About

Jean-Charles Bazin is a leading researcher in computer vision and robotics, whose work has fundamentally advanced how autonomous systems perceive and navigate their environment. His primary research areas include omnidirectional vision, rotation estimation, and geometric scene understanding, with a particular focus on urban environments. Bazin’s most impactful contribution is the development of the "3-line RANSAC" algorithm for orthogonal vanishing point detection, a seminal paper with 105 citations that provides a robust, vision-based alternative to traditional GPS and inertial sensors for estimating a robot’s orientation. This work, along with his highly cited 2012 study on rotation estimation using omnidirectional vision (87 citations), has become foundational for automatic control, stabilization, and 3D reconstruction in ground and aerial vehicles. Bazin has also pioneered techniques for extracting geometric primitives like rectangles from catadioptric images, addressing previously underexplored challenges in robotic vision. His innovative approach to improving feature matching by fusing gyroscope data with visual information further demonstrates his commitment to enhancing the reliability of motion and localization estimation. Through these achievements, Bazin has established himself as a key figure in making vision-based navigation more accurate and resilient for real-world robotic applications.