G. Radonich

About

G. Radonich’s research lies at the intersection of computer vision and robotics, with a primary focus on optical flow—the pattern of apparent motion of objects in a visual scene. In two foundational papers, Radonich developed algorithms to recover both motion and depth from optical flow, addressing a core challenge in autonomous navigation: how a robot can understand its own movement and the three-dimensional structure of its environment using only passive vision from onboard cameras. This work, published in the late 1980s and early 1990s, proposed methods that allow robots to interpret visual input without relying on active sensors like lidar, a concept that remains relevant in modern autonomous systems. While the papers have accrued modest citation counts—two each—they represent early, principled contributions to a field that has since exploded in importance. Radonich’s emphasis on passive vision as a viable sensor modality for robotics helped lay conceptual groundwork for later advances in visual odometry and structure-from-motion. For students and researchers exploring the history of robotic perception, these papers offer a clear window into the foundational thinking that preceded today’s deep learning-driven approaches.

Research Focus

Key Achievements

Top Papers

1. 1
   An Algorithm for Optical Flow

   2 citations · 1990
2. 2
   The Recovery of Motion and Depth from Optical Flow

   2 citations · 1989

Key Collaborators

Contact & Links