公开(公告)号：US12106531B2

公开(公告)日：2024-10-01

申请号：US17383362

申请日：2021-07-22

Applicant: Microsoft Technology Licensing, LLC

Inventor： Lijuan Wang ,  Zicheng Liu ,  Ying Jin ,  Hongli Deng ,  Kun Luo ,  Pei Yu ,  Yinpeng Chen

IPC: G06V40/10 ,  G06T7/70 ,  G06V10/22

CPC classification number: G06V10/22 ,  G06T7/70 ,  G06V40/10 ,  G06T2207/30196

Abstract: To improve the accuracy and efficiency of object detection through computer digital image analysis, the detection of some objects can inform the sub-portion of the digital image to which subsequent computer digital image analysis is directed to detect other objects. In such a manner object detection can be made more efficient by limiting the image area of a digital image that is analyzed. Such efficiencies can represent both computational efficiencies and communicational efficiencies arising due to the smaller quantity of digital image data that is analyzed. Additionally, the detection of some objects can render the detection of other objects more accurate by adjusting confidence thresholds based on the detection of those related objects. Relationships between objects can be utilized to inform both the image area on which subsequent object detection is performed and the confidence level of such subsequent object detection.

公开(公告)号：US11036975B2

公开(公告)日：2021-06-15

申请号：US16220797

申请日：2018-12-14

Applicant: Microsoft Technology Licensing, LLC

Inventor： Noranart Vesdapunt ,  Baoyuan Wang ,  Ying Jin ,  Pierrick Arsenault

IPC: G06K9/00 ,  G06T7/73 ,  G06F3/01 ,  G06N3/08 ,  G06K9/62 ,  G06T1/20

Abstract: Described herein is a human pose prediction system and method. An image comprising at least a portion of a human body is received. A trained neural network is used to predict one or more human features (e.g., joints/aspects of a human body) within the received image, and, to predict one or more human poses in accordance with the predicted one or more human features. The trained neural network can be an end-to-end trained, single stage deep neural network. An action is performed based on the predicted one or more human poses. For example, the human pose(s) can be displayed as an overlay with received image.