Title :
Self-calibration of a camera from video of a walking human
Author :
Lv, Fengjun ; Zhao, Tao ; Nevatia, Ram
Author_Institution :
Inst. for Robotics & Intelligent Syst., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
Analysis of human activity from a video camera is simplified by the knowledge of the camera´s intrinsic and extrinsic parameters. We describe a technique to estimate such parameters from image observations without requiring measurements of scene objects. We first develop a general technique for calibration using vanishing points and vanishing line. We then describe a method for estimating the needed points and line by observing the motion of a human in the scene. Experimental results, including error estimates, are presented.
Keywords :
calibration; image motion analysis; image sequences; matrix algebra; parameter estimation; video cameras; error estimates; extrinsic parameters; human activity analysis; image observations; intrinsic parameters; self-calibration; vanishing line; vanishing points; video camera; walking human; Calibration; Cameras; Carbon capture and storage; Charge-coupled image sensors; Humans; Intelligent robots; Intelligent systems; Layout; Legged locomotion; Robot vision systems;
Conference_Titel :
Pattern Recognition, 2002. Proceedings. 16th International Conference on
Print_ISBN :
0-7695-1695-X
DOI :
10.1109/ICPR.2002.1044793