公开(公告)号：US09965717B2

公开(公告)日：2018-05-08

申请号：US14940916

申请日：2015-11-13

Applicant: ADOBE SYSTEMS INCORPORATED

Inventor： Zhaowen Wang ,  Xianming Liu ,  Hailin Jin ,  Chen Fang

IPC: G06K9/00 ,  G06N3/04 ,  G06K9/62 ,  G06K9/46 ,  G06N3/08 ,  G06Q50/00

CPC classification number: G06N3/0454 ,  G06K9/4628 ,  G06K9/6257 ,  G06K9/627 ,  G06K9/628 ,  G06K9/6284 ,  G06K9/629 ,  G06N3/04 ,  G06N3/08 ,  G06Q50/01

Abstract: Embodiments of the present invention relate to learning image representation by distilling from multi-task networks. In implementation, more than one single-task network is trained with heterogeneous labels. In some embodiments, each of the single-task networks is transformed into a Siamese structure with three branches of sub-networks so that a common triplet ranking loss can be applied to each branch. A distilling network is trained that approximates the single-task networks on a common ranking task. In some embodiments, the distilling network is a Siamese network whose ranking function is optimized to approximate an ensemble ranking of each of the single-task networks. The distilling network can be utilized to predict tags to associate with a test image or identify similar images to the test image.