Understanding people motion in video sequences using Voronoi diagrams

Crowd Analysis Using Computer Vision Techniques

“…This area represents a comfort zone during interpersonal communication, and can be reduced in specific environments (crowded environments). The classification of interpersonal relationships proposed by Hall (intimate, personal, social, and public) can be explored by crowd analysis algorithms, as done in [2].…”

“…To overcome this problem, some authors [41], [2] have used crowd simulation algorithms to generate controlled situations with known ground truth to test their algorithms. In fact, concepts related to crowd simulation are also being explored to distinguish normal and abnormal behaviors, as in [42].…”

“…In [2], an algorithm for group detection and classification as voluntary or involuntary based on computer vision was proposed. In their work, a top-down camera is used to track individuals through, and Voronoi diagrams are used to quantify the sociological concept of personal space.…”

“…The temporal evolution of the Voronoi diagrams is used to identify groups in the scene, and the portion of the personal space within each individual's field of view is used to classify the groups as voluntary or involuntary. An interesting aspect of [2] is the use of sociological aspects to analyze crowds, but one drawback is the need of tracking each individual in the scene, which may be difficult in denser scenes.…”

“…In a high-level analysis, one could segment their body parts (e.g., head, face, hands, and arms) that could be used in gesture recognition or machine-human interaction, for example. It is also possible to deal with group behavior, once people can be part of high-level structures, such as groups or crowds [2], [3]. In particular, the behavioral analysis of crowded scenes is of great interest in a large number of applications [4], such as Crowd Management ■ : It can be used for developing crowd management strategies, to avoid crowd related disasters and insure public safety.…”