| 摘要 |
METHOD FOR VIDEO OBJECT DETECTION Abstract Disclosed herein are a system and method for performing foreground/background separation on an input image. The method performs the following steps for each input visual element in the input image. The method identifies a corresponding model visual element in a scene model (200) associated with the input image, the model visual element 5 being associated with a set of element models (210), each element model (220) including a plurality of visual data sets (230, 240). The method then selects an element model (220) from the set of element models (210), dependent upon a visual distance between the input visual element and a visual data set (230, 240) of the selected element model (220) satisfying a predetermined criterion. The method classifies the input visual element as one io of foreground and background, dependent upon the selected element model (220), and then updates each visual data set (230, 240) in the selected element model (220), dependent upon the input visual element and at least first and second different methods of updating a visual data set. iA4Q4t I nloC IRN: 958464 |
