发明名称 Method and arrangement for image retrieval based on multiple images
摘要 A Method for retrieving at least one image from a database (DB) of images based on at least two input images (image i, image j), comprises the steps of determining (120) first low level feature correspondences (LC1) between said at least two input images (image i, image j), searching (200) within said database (DB) for at least two sets (Mi, Mj) of images respectively matching said at least to images (image i, image j), determining (120) second low level features correspondences (LC2) between respective images from said at least two sets of images (Mi, Mj), determining (130) a first set of relationships (RLC1) between entities of said at least two input images (image i, image j) based on said first low level feature correspondences (LC1), determining (130) a second set of relationships (RLC2) between respective entities of said respective images from said at least two sets of images (Mi, Mj) based on said second low level feature correspondences (LC2), identifying (300) matching relationships between said first set of relationships (RLC1) and said second set of relationships (RLC2), checking (400) the quality of the matching relationships, and, if matching relationships of sufficient quality are found, retrieving (500) the at least one image corresponding to the matching relationships of said second set from said database, and providing said at least one image of said second set as output.
申请公布号 US9535929(B2) 申请公布日期 2017.01.03
申请号 US201314653120 申请日期 2013.12.17
申请人 Alcatel Lucent 发明人 Namboodiri Vinay;Feki Mohamed Ali;Six Erwin
分类号 G06K9/68;G06F17/30;G06K9/62 主分类号 G06K9/68
代理机构 Fay Sharpe, LLP 代理人 Fay Sharpe, LLP
主权项 1. A computer-implemented method for retrieving at least one image from a database of images based on at least two input images, said method comprising: determining, by one or more processors, first low level feature correspondences between said at least two input images; searching the one or more processors within said database for at least two sets of images respectively matching said at least two images; determining, by the one or more processors, second low level features correspondences between respective images from said at least two sets of images; determining, by the one or more processors, a first set of relationships between entities of said at least two input images based on said first low level feature correspondences; determining, by the one or more processors, a second set of relationships between respective entities of said respective images from said at least two sets of images based on said second low level feature correspondences and attributing a weighting score to said relationships; identifying in the one or more processors matching relationships between said first set of relationships and said second set of relationships; checking, in the one or more processors, the quality of the matching relationships, by checking their weighting score and, if matching relationships of sufficient quality are found, as expressed by a high weighting score against a threshold, retrieving the at least one image corresponding to the matching relationships of said second set from said database; and providing by the one or more processors said at least one image of said second set as output.
地址 Boulogne-Billancourt FR