发明名称 MULTI-CUE OBJECT ASSOCIATION
摘要 Multiple discrete objects within a scene image captured by a single camera track are distinguished as un-labeled from a background model within a first frame of a video data input. Object position, object appearance and/or object size attributes are determined for each of the blobs, and costs determined to assign to existing blobs of existing object tracks as a function of the determined attributes. The un-labeled object blob that has a lowest cost of association with any of the existing object tracks is labeled with the label of that track having the lowest cost, said track is removed from consideration for labeling remaining un-labeled object blobs, and the process iteratively repeated until each of the track labels have been used to label one of the un-labeled blobs.
申请公布号 US2015023560(A1) 申请公布日期 2015.01.22
申请号 US201414506974 申请日期 2014.10.06
申请人 International Business Machines Corporation 发明人 Datta Ankur;Feris Rogerio S.;Pankanti Sharathchandra U.;Zhai Yun
分类号 G06K9/00;G06T7/00;G06T7/20 主分类号 G06K9/00
代理机构 代理人
主权项 1. A method to track multiple discrete objects within a scene image captured by a single camera, the method comprising: distinguishing a plurality of foreground object bounding box blobs from a background model within a first frame of a video data input from a single camera view via a background subtraction process; initializing each of the plurality of foreground object bounding box blobs as un-labeled; determining at least one image attribute for each of the un-labeled object blobs; determining at least one attribute cost for assigning each of the un-labeled object blobs to existing blobs of each of at least one existing object track determined by performing a background subtraction process on a previous frame of the video data input; labeling one of the un-labeled object blobs that has a lowest cost of association with any of the existing object tracks with a label of one of the existing tracks that has the lowest cost of association; removing the one track of the label of the labeled object blob from consideration for labeling remaining others of the un-labeled object blobs; and iteratively repeating steps of labeling each of the remaining ones of the un-labeled object blobs that each have a lowest cost of association with any of the existing object tracks of the remaining ones of the existing tracks that have not been removed for consideration, and removing the tracks of the labels of the labeled object blobs from consideration for labeling of remaining others of the un-labeled object blobs, until the labels of each of the tracks have been used to label ones of the un-labeled object blobs.
地址 Armonk NY US