Visual 3D target tracking for autonomous vehicles

In this paper, an algorithim is proposed to ideentify and tracking moving objects for the autonomous vehicles target following application. It is a difficult problem because both the targets and the caiyieras are inoving. Hcre optical flow fields, color features, stereo pair disparity are riscd as visual features while vehicles’ motion sensors are used to detcrminc the camera motion. Then this paper proposes a data fusion algoritlirii which integrates inforniatioii obtained from different visual cues and thc caniera riwtion seiisor data. The fusion algorithm dcterinines the speed and relative position of the interested target in the 3D world coordinate for the vehicle to track. This paper presents a detailed description of the thee-dimensional (3D) target tracking algorithm using an Extended Kalriian Filter. Experiniental results arc presented to demonstrate the performance of the proposed scheirie using different image sequences.