发明名称 Exemplar-based feature weighting
摘要 In an example embodiment, for each of the image exemplars, a first location offset between an actual landmark location for a first landmark in the image exemplar and a predicted landmark location for the first landmark in the image exemplar is determined. Then, a probability that the image recognition process applied using the first feature produces an accurate identification of the first landmark in the image exemplars is determined based on the first location offsets for each of the image exemplars. A weight may then be assigned to the first feature based on the derived probability. An image recognition process may then be performed on an image, the image recognition process utilizing a voting process, for each of one or more features, for one or more landmarks in the plurality of image exemplars, the voting process for the first feature weighted according to the weight assigned to the first feature.
申请公布号 US9129152(B2) 申请公布日期 2015.09.08
申请号 US201314080010 申请日期 2013.11.14
申请人 ADOBE SYSTEMS INCORPORATED 发明人 Brandt Jonathan;Lin Zhe;Smith Brandon M.
分类号 G06K9/00;G06K9/62 主分类号 G06K9/00
代理机构 Shook, Hardy & Bacon L.L.P. 代理人 Shook, Hardy & Bacon L.L.P.
主权项 1. A method of performing image recognition using a plurality of image exemplars, each of the image exemplars having a plurality of features, the method comprising: for each of the image exemplars, determining a first location offset between an actual landmark location for a first landmark in the image exemplar and a predicted landmark location for the first landmark in the image exemplar, the predicted landmark location derived using an image recognition process using a first feature of the image exemplar, wherein the first location offset for each of the image exemplars is relative to a size of a face in each of the image exemplars; deriving a probability that the image recognition process applied using the first feature produces an accurate identification of the first landmark in the image exemplars based on the first location offsets for each of the image exemplars; assigning a weight to the first feature based on the derived probability; and performing the image recognition process on an image, the image recognition process utilizing a voting process, for each of one or more features, for one or more landmarks in the plurality of image exemplars, the voting process for the first feature weighted according to the weight assigned to the first feature.
地址 San Jose CA US