A Simplified Error Model for Height Estimation Using a Single Camera

Abstract

Intelligent solutions for video surveillance require reliable object features for robust tracking. Surveillance cameras are generally low resolution and hence impose restrictions on the ability to identify sufficient features and descriptors for effective tracking. In such applications, reliable estimate of the object height from a single uncalibrated camera can prove to be a very useful feature. Several approaches have been developed to obtain height information of an object from image sequences, but are limited in their utility due to errors arising from varying image and object properties. In this paper, we present a simplified error model dependent on the location of the object in the image and the estimated camera height. Further, the developed model is used to obtain reliable height estimates of objects in different image sequences and the result is compared to estimates obtained without the use of the model. Significant reduction in error associated with the measurement is observed while making the estimate independent of camera and object variations. We then demonstrate its utility in object tracking.