Principal axis-based correspondence between multiple cameras for people tracking

• Published in: IEEE transactions on pattern analysis and machine intelligence 2006-04, Vol.28 (4), p.663-671
• Main Authors: Weiming Hu, Min Hu, Xue Zhou, Tieniu Tan, Jianguang Lou, Maybank, S.
• Format: Article
• Language: English
• Subjects: Algorithms; Applied sciences; Artificial Intelligence; BASIC (programming language); Calibration; Cameras; Computer science; control theory; systems; Computer vision; Correspondence between multiple cameras; Exact sciences and technology; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image sequences; Imaging, Three-Dimensional - methods; Information Storage and Retrieval - methods; Models, Biological; Monitoring; Motion detection; Movement - physiology; Noise robustness; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; people tracking; Photogrammetry - methods; principal axes; Reproducibility of Results; Robustness; Sensitivity and Specificity; Similarity; Studies; Subtraction Technique; Surveillance; Tracking; Video Recording - methods; Video sequences; Visual; Working environment noise
• ISSN: 0162-8828; 1939-3539
• DOI: 10.1109/TPAMI.2006.80

Visual surveillance using multiple cameras has attracted increasing interest in recent years. Correspondence between multiple cameras is one of the most important and basic problems which visual surveillance using multiple cameras brings. In this paper, we propose a simple and robust method, based on principal axes of people, to match people across multiple cameras. The correspondence likelihood reflecting the similarity of pairs of principal axes of people is constructed according to the relationship between "ground-points" of people detected in each camera view and the intersections of principal axes detected in different camera views and transformed to the same view. Our method has the following desirable properties; 1) camera calibration is not needed; 2) accurate motion detection and segmentation are less critical due to the robustness of the principal axis-based feature to noise; 3) based on the fused data derived from correspondence results, positions of people in each camera view can be accurately located even when the people are partially occluded in all views. The experimental results on several real video sequences from outdoor environments have demonstrated the effectiveness, efficiency, and robustness of our method.