公开(公告)号：US11947589B2

公开(公告)日：2024-04-02

申请号：US17710761

申请日：2022-03-31

Applicant: Microsoft Technology Licensing, LLC

Inventor： Li Huang ,  Rui Xia ,  Zhiting Chen ,  Kun Wu ,  Meenaz Merchant ,  Kamal Ginotra ,  Arun K. Sacheti ,  Chu Wang ,  Andrew Lawrence Stewart ,  Hanmu Zuo ,  Saurajit Mukherjee

IPC: G06F16/50 ,  G06F16/532 ,  G06F16/535 ,  G06F16/56

CPC classification number: G06F16/535 ,  G06F16/532 ,  G06F16/56

Abstract: Systems and methods directed to returning personalized image-based search results are described. In examples, a query including an image may be received, and a personalized item embedding may be generated based on the image and user profile information associated with a user. Further, a plurality of candidate images may be obtained based on the personalized item embedding. The candidate images may then be ranked according to a predicted level of user engagement for a user, and then diversified to ensure visual diversity among the ranked images. A portion of the diversified images may then be returned in response to an image-based search.

公开(公告)号：US20200019628A1

公开(公告)日：2020-01-16

申请号：US16036224

申请日：2018-07-16

Applicant: Microsoft Technology Licensing, LLC

Inventor： Xi Chen ,  Houdong Hu ,  Li Huang ,  Jiapei Huang ,  Arun Sacheti ,  Linjun Yang ,  Rui Xia ,  Kuang-Huei Lee ,  Meenaz Merchant ,  Sean Chang Culatana

IPC: G06F17/30 ,  G06K9/62 ,  G06N5/04 ,  G06N99/00 ,  G06F17/11

Abstract: Representative embodiments disclose mechanisms to perform visual intent classification or visual intent detection or both on an image. Visual intent classification utilizes a trained machine learning model that classifies subjects in the image according to a classification taxonomy. The visual intent classification can be used as a pre-triggering mechanism to initiate further action in order to substantially save processing time. Example further actions include user scenarios, query formulation, user experience enhancement, and so forth. Visual intent detection utilizes a trained machine learning model to identify subjects in an image, place a bounding box around the image, and classify the subject according to the taxonomy. The trained machine learning model utilizes multiple feature detectors, multi-layer predictions, multilabel classifiers, and bounding box regression.