Open AccessMOVING HUMAN PATH TRACKING BASED ON VIDEO SURVEILLANCE IN 3D INDOOR SCENARIOS
Author(s) -
Yan Zhou,
Sisi Zlatanova,
Zhe Wang,
Yeting Zhang,
Liu Liu
Publication year - 2016
Publication title -
isprs annals of the photogrammetry, remote sensing and spatial information sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.356
H-Index - 38
eISSN - 2194-9042
pISSN - 2196-6346
DOI - 10.5194/isprsannals-iii-4-97-2016
Subject(s) - computer science , robustness (evolution) , computer vision , artificial intelligence , path (computing) , process (computing) , real time computing , tracking (education) , computer network , psychology , pedagogy , biochemistry , chemistry , gene , operating system
Video surveillance systems are increasingly used for a variety of 3D indoor applications. We can analyse human behaviour, discover and avoid crowded areas, monitor human traffic and so forth. In this paper we concentrate on use of surveillance cameras to track and reconstruct the path a person has followed. For the purpose we integrated video surveillance data with a 3D indoor model of the building and develop a single human moving path tracking method. We process the surveillance videos to detected single human moving traces; then we match the depth information of 3D scenes to the constructed 3D indoor network model and define the human traces in the 3D indoor space. Finally, the single human traces extracted from multiple cameras are connected with the help of the connectivity provided by the 3D network model. Using this approach, we can reconstruct the entire walking path. The provided experiments with a single person have verified the effectiveness and robustness of the method.