公开(公告)号：US11636694B2

公开(公告)日：2023-04-25

申请号：US17185800

申请日：2021-02-25

Applicant: QUALCOMM Incorporated

Inventor： Xin Li ,  Chun-Ting Huang ,  Lei Wang ,  Ning Bi

IPC: G06K9/62 ,  G06V20/59 ,  G06V20/40 ,  G06F18/214 ,  G06F18/20

Abstract: Systems and techniques are provided for performing video-based activity recognition. For example, a process can include extracting, using a first machine learning model, first one or more features from a first frame and second one or more features from a second frame. The first one or more features and the second one or more features are associated with a person driving a vehicle. The process can include processing, using a second machine learning model, the first one or more features and the second one or more features. The process can include determining, based on processing of the first one or more features and the second one or more features using the second machine learning model, at least one activity associated with the person driving the vehicle.

公开(公告)号：US11188740B2

公开(公告)日：2021-11-30

申请号：US16719900

申请日：2019-12-18

Applicant: QUALCOMM incorporated

Inventor： Chun-Ting Huang ,  Lei Wang ,  Zhen Wang ,  Xiaoliang Bai ,  Ning Bi

IPC: G06K9/00 ,  G06T7/73 ,  G06K9/46 ,  G06K9/62

Abstract: Methods, systems, and devices for object detection are described. A device may receive an image, and detect, via a first stage of a cascade neural network, object recognition information over one or more angular orientations during a first pass. The device may determine, via a second stage of the cascade neural network, a confidence score associated with one or more of the candidate object in the image, the candidate bounding box associated with the candidate object in the image, or one or more object features of the candidate object in the image, or an orientation of the candidate object in the image, or a combination thereof. The device may identify, via a third stage of the cascade neural network, whether to detect the object recognition information during a second pass based on the confidence score satisfying a threshold.

公开(公告)号：US12142084B2

公开(公告)日：2024-11-12

申请号：US17561309

申请日：2021-12-23

Applicant: QUALCOMM INCORPORATED

Inventor： Chun-Ting Huang ,  Lei Wang ,  Ning Bi

IPC: G06V40/40 ,  G06V10/82 ,  G06V40/16

Abstract: Methods, systems, and apparatuses are provided to automatically determine whether an image is spoofed. For example, a computing device may obtain an image, and may execute a trained convolutional neural network to ingest elements of the image. Further, and based on the ingested elements of the image, the executed trained convolutional neural network generates an output map that includes a plurality of intensity values. In some examples, the trained convolutional neural network includes a plurality of down sampling layers, a plurality of up sampling layers, and a plurality of joint spatial and channel attention layers. Further, the computing device may determine whether the image is spoofed based on the plurality of intensity values. The computing device may also generate output data based on the determination of whether the image is spoofed, and may store the output data within a data repository.

公开(公告)号：US20210192756A1

公开(公告)日：2021-06-24

申请号：US16719062

申请日：2019-12-18

Applicant: QUALCOMM Incorporated

Inventor： Chun-Ting Huang ,  Lei Wang ,  Ning Bi ,  Alex Jong

IPC: G06T7/246 ,  G06T7/238 ,  G06T7/215 ,  G06T7/73 ,  G06N3/04

Abstract: Methods, systems, and devices for image processing are described. A device may receive a first frame including a candidate object. The device may detect first object recognition information based on the first frame or a portion of the first frame. The first object recognition information may include the candidate object or a first candidate bounding box associated with the candidate object. The device may detect second object recognition information based on the first object recognition information, a second frame, or a portion of the second frame. The second object recognition information may include the candidate object in the second frame, a second candidate bounding box associated with the candidate object, or features of the candidate object. The device may estimate motion information associated with the candidate object in the first frame, and track the candidate object in the second frame based on the motion information.

公开(公告)号：US20210192182A1

公开(公告)日：2021-06-24

申请号：US16719900

申请日：2019-12-18

Applicant: QUALCOMM Incorporated

Inventor： Chun-Ting Huang ,  Lei Wang ,  Zhen Wang ,  Xiaoliang Bai ,  Ning Bi

IPC: G06K9/00 ,  G06T7/73 ,  G06K9/62 ,  G06K9/46

Abstract: Methods, systems, and devices for object detection are described. A device may receive an image, and detect, via a first stage of a cascade neural network, object recognition information over one or more angular orientations during a first pass. The device may determine, via a second stage of the cascade neural network, a confidence score associated with one or more of the candidate object in the image, the candidate bounding box associated with the candidate object in the image, or one or more object features of the candidate object in the image, or an orientation of the candidate object in the image, or a combination thereof. The device may identify, via a third stage of the cascade neural network, whether to detect the object recognition information during a second pass based on the confidence score satisfying a threshold.