| 摘要 |
The field of the invention relates to a computer vision process in which video and image analysis is used to detect and track objects, and to generate events that are associated with changes in the objects, such as changes in the objects presence, trajectory or pose. The invention could find application in: home automation (e.g. controlling lighting), surveillance, security (e.g. determining whether a person has left baggage unattended), and marketing or commercial environments (e.g. monitoring people entering and leaving a shop). In the claimed process a track record (or trajectory) for one or more detected objects is created, the track record defining metadata or parameters for that detected object; and that track record is then linked to one or more higher-level track records for one or more higher-level objects. For example, tracks for hands 1008 or faces 1005 could be linked to a higher-level track 1011 for a person 1007. Decisions about whether to link a hand to a person are influenced by, for example, the distance between the hand and the person (see Fig. 12). |
