Tracking across multiple cameras with disjoint views
US7450735B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Oct 16, 2004 |
| Grant date | Nov 11, 2008 |
| Priority date | — |
| Expiry date | May 18, 2026 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/10016
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Tracking and surveillance methods and systems for monitoring objects passing in front of non-overlapping cameras. Invention finds corresponding tracks from different cameras and works out which object passing in front of the camera(s) made the tracks, in order to track the object from camera to camera. The invention uses an algorithm to learn inter-camera spatial temporal probability using Parzen windows, learns inter-camera appearance probabilities using distribution of Bhattacharyya distances between appearance models, establishes correspondences based on Maximum A Posteriori (MAP) framework combining both spatial temporal and appearance probabilities, and updates learned probabilities throughout the lifetime of the system.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.