Mahdi Karimian

Papers

Total Citations

H-Index

About

Mahdi Karimian’s research focuses on advancing autonomous navigation through computer vision and sensor fusion, with a particular emphasis on robust rotation estimation for self-driving cars and mobile robots. His major contribution lies in developing efficient, vision-based methods for determining vehicle orientation when GPS signals are unreliable or unavailable. In his highly cited work “Online visual gyroscope for autonomous cars” (2016, 7 citations), Karimian proposed a novel approach that uses visual data alone to estimate rotation in real time, addressing critical limitations of inertial sensors like drift and calibration errors. He further refined this technique in “3-Point RANSAC for fast vision based rotation estimation using GPU technology” (2016, 3 citations), where he leveraged GPU acceleration to dramatically speed up the RANSAC algorithm by decoupling rotation from translation estimation—making it far more efficient for real-world deployment. Karimian’s work is notable for its practical focus on overcoming real-world navigation challenges, offering computationally lightweight solutions that enhance the reliability of autonomous systems. His research has direct implications for improving the safety and autonomy of self-driving cars in GPS-denied environments such as tunnels, urban canyons, and dense forests.