Max Robotics — Global Service Robot Directory + US Certification

About

Peter Sturm has made foundational contributions to computer vision, particularly in camera calibration, 3D reconstruction, and vision for robotics. His key research areas include omnidirectional camera calibration, uncalibrated 3D vision, and real-time UAV perception. Sturm’s most cited work, "Calibration of omnidirectional cameras in practice: A comparison of methods" (2011, 84 citations), provides a comprehensive evaluation of calibration techniques for wide-field-of-view cameras, becoming a standard reference for researchers and practitioners. His PhD thesis (1997, 32 citations) advanced the theory of projective reconstruction and critical motion sequences for auto-calibration, addressing the practical challenge of 3D reconstruction without offline camera calibration—a problem critical for dynamic applications. Sturm also pioneered vision-based attitude and altitude estimation for UAVs in dark environments (2011, 15 and 11 citations), integrating fisheye cameras with laser projectors to enable robust flight control under low light. Additionally, his work on detecting semi-transparent obstacles using collective-reward approaches (2011, 12 and 8 citations) tackles the underexplored problem of non-opaque object perception, with implications for safe robot navigation in man-made environments. With over 160 citations across his top papers, Sturm’s research bridges theoretical rigor and practical deployment, influencing both academic computer vision and real-world autonomous systems.