In advanced video surveillance systems, people localization is usually a part of the complete system, and should be accomplished in short time so as to reserve sufficient processing time for subsequent high-level analysis, such as abnormal event/behavior detection, intruder detection, etc. Hence, in addition to localization accuracy, computational efficiency is of critical importance as well. In our method, we adopt a vanishing point-based line sampling scheme, and propose a fast multi- camera people localization approach capable of locating a crowd of dense people and estimating their heights in fairly short time with high accuracy.
(a) Segmented foreground regions (in teal) and 2D line samples (in red). (b) Localization results displayed by 3D major axes. (c) Localization results illustrated with 3D bounding boxes.