Alex Teichman

University of Pennsylvania, Laboratoire d'Informatique de Paris-Nord, Stanford University

Papers

Total Citations

421

H-Index

About

Alex Teichman’s research lies at the intersection of robotics, computer vision, and machine learning, with a focus on enabling autonomous systems to perceive and interact with the world. His most influential work, “Towards 3D object recognition via classification of arbitrary object tracks” (176 citations), pioneered a tracking-based approach to object recognition for autonomous vehicles—classifying entire object trajectories rather than static point clouds, a paradigm shift that improved robustness in dynamic environments. He extended this in “Practical object recognition in autonomous driving and beyond” (58 citations), outlining the challenges and solutions for deploying recognition systems in real-world autonomous taxis. Earlier, Teichman contributed to modular robotics with “Automatic Configuration Recognition Methods in Modular Robots” (111 citations), developing algorithms to identify and control reconfigurable robot morphologies. He also advanced unsupervised learning with “Exponential family sparse coding with applications to self-taught learning” (71 citations), showing how sparse coding could leverage unlabeled data to improve supervised tasks. His later work on “Group induction” (5 citations) aimed to reduce the annotation burden in machine perception. Teichman’s contributions have been instrumental in bridging perception and autonomy, with his object tracking methods directly influencing modern autonomous driving systems.