TY - CHAP T1 - Video Mensuration Using a Stationary Camera T2 - Computer Vision – ECCV 2006Computer Vision – ECCV 2006 Y1 - 2006 A1 - Guo,Feng A1 - Chellapa, Rama ED - Leonardis,Aleš ED - Bischof,Horst ED - Pinz,Axel AB - This paper presents a method for video mensuration using a single stationary camera. The problem we address is simple, i.e., the mensuration of any arbitrary line segment on the reference plane using multiple frames with minimal calibration. Unlike previous solutions that are based on planar rectification, our approach is based on fitting the image of multiple concentric circles on the plane. Further, the proposed method aims to minimize the error in mensuration. Hence we can calculate the mensuration of the line segments not lying on the reference plane. Using an algorithm for detecting and tracking wheels of an automobile, we have implemented a fully automatic system for wheel base mensuration. The mensuration results are accurate enough that they can be used to determine the vehicle classes. Furthermore, we measure the line segment between any two points on the vehicle and plot them in top and side views. JA - Computer Vision – ECCV 2006Computer Vision – ECCV 2006 T3 - Lecture Notes in Computer Science PB - Springer Berlin / Heidelberg VL - 3953 SN - 978-3-540-33836-9 UR - http://dx.doi.org/10.1007/11744078_13 ER -