Object recognition method and apparatus

    公开(公告)号:US12217477B2

    公开(公告)日:2025-02-04

    申请号:US17680668

    申请日:2022-02-25

    Abstract: In an object recognition method, an object recognition device obtains AER data of a to-be-recognized object, wherein the AER data includes a plurality of AER events of the to-be-recognized object, each AER event comprising a timestamp and address information. The object recognition device extracts a plurality of feature maps of the AER data. Each feature map including partial spatial information and partial temporal information of the to-be-recognized object, and the partial spatial information and the partial temporal information are obtained based on the timestamp and the address information of each AER event. The object recognition device then recognizes the to-be-recognized object based on the plurality of feature maps of the AER data.

    METHOD AND COMPUTER SYSTEM FOR TRAINING A NEURAL NETWORK MODEL

    公开(公告)号:US20240095531A1

    公开(公告)日:2024-03-21

    申请号:US18521763

    申请日:2023-11-28

    CPC classification number: G06N3/08

    Abstract: A method trains a neural network model in a computer system. The neural network model includes one or more layers each including one or more neurons. The one or more layers include at least one first layer and one last layer. Each neurons are configured to perform forward propagation of one or more input values by applying weights to the one or more input values and generating an output value based on a function applied to the sum of the weighted input values. The neurons of any given layer, but the last layer, of the one or more layers are connected with the one or more neurons of a consecutive layer. The neurons of any given layer, but the first layer, of the one or more layers are connected with the one or more neurons of a preceding layer.

    Camera Motor, Camera Module, and Electronic Device

    公开(公告)号:US20230007153A1

    公开(公告)日:2023-01-05

    申请号:US17782038

    申请日:2020-10-28

    Abstract: A camera motor includes a motor housing, a lens bearing apparatus, a cushioning part, and a drive apparatus. The motor housing has an accommodating cavity, and the lens bearing apparatus is accommodated in the accommodating cavity and is spaced apart from an inner wall of the accommodating cavity. The cushioning part is disposed on an outer surface of the lens bearing apparatus. The drive apparatus is accommodated in the accommodating cavity, and the drive apparatus can drive the lens bearing apparatus to move in the accommodating cavity.

    Image processing method and apparatus, and storage medium

    公开(公告)号:US12299911B2

    公开(公告)日:2025-05-13

    申请号:US18064144

    申请日:2022-12-09

    Abstract: An image processing method and apparatus, and a storage medium are provided, and relate to the image processing field. A dependency relationship between features of texture primitives of an image may be obtained based on direction information and a multi-scale feature map of the image, at least one group of texture features of the image may be obtained based on a feature map of the image on at least one scale, and a texture representation result of the image may be obtained based on the dependency relationship and the at least one group of texture features. Then, the image may be processed based on the texture representation result of the image. Because the texture representation result of the image can reflect more perfect texture information of the image, an image processing effect is better when image processing such as image recognition, image segmentation, or image synthesis is performed.

    OBJECT RECOGNITION METHOD AND APPARATUS

    公开(公告)号:US20220180619A1

    公开(公告)日:2022-06-09

    申请号:US17680668

    申请日:2022-02-25

    Abstract: In an object recognition method, an object recognition device obtains AER data of a to-be-recognized object, wherein the AER data includes a plurality of AER events of the to-be-recognized object, each AER event comprising a timestamp and address information. The object recognition device extracts a plurality of feature maps of the AER data. Each feature map including partial spatial information and partial temporal information of the to-be-recognized object, and the partial spatial information and the partial temporal information are obtained based on the timestamp and the address information of each AER event. The object recognition device then recognizes the to-be-recognized object based on the plurality of feature maps of the AER data.

    Camera module and terminal
    9.
    发明授权

    公开(公告)号:US11089189B2

    公开(公告)日:2021-08-10

    申请号:US16642309

    申请日:2017-09-12

    Abstract: A camera module includes a lens and a voice coil motor (VCM), and the lens is placed in the VCM, where a first portion of an inner side of the VCM is attached to a first portion of an outer side of the lens, and a second portion of the inner side of the VCM is provided with a screw thread structure recessed relative to the first portion of the inner side of the VCM. The screw thread structure and a second portion of the outer side of the lens form a groove portion, and the groove portion is filled with adhesive. The first portion of the VCM is a lower portion, and the second portion of the VCM is an upper portion. The first portion of the lens is a lower portion, and the second portion of the lens is an upper portion.

Patent Agency Ranking