-
公开(公告)号：US20230206698A1
公开(公告)日：2023-06-29
申请号：US17561309
申请日：2021-12-23
Applicant: QUALCOMM INCORPORATED
Inventor： Chun-Ting HUANG , Lei Wang , Ning Bi
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
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
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：US10706267B2
公开(公告)日：2020-07-07
申请号：US15869342
申请日：2018-01-12
Applicant: QUALCOMM Incorporated
Inventor： Lei Wang , Ning Bi , Yingyong Qi
Abstract: Methods, systems, and devices for object recognition are described. Generally, the described techniques provide for a compact and efficient convolutional neural network (CNN) model for facial recognition. The proposed techniques relate to a light model with a set of layers of convolution and one fully connected layer for feature representation. A new building block of for each convolution layer is proposed. A maximum feature map (MFM) operation may be employed to reduce channels (e.g., by combining two or more channels via maximum feature selection within the channels). Depth-wise separable convolution may be employed for computation reduction (e.g., reduction of convolution computation). Batch normalization may be applied to normalize the output of the convolution layers and the fully connected layer (e.g., to prevent overfitting). The described techniques provide a compact and efficient CNN model which can be used for efficient and effective face recognition.
-
公开(公告)号：US20190205694A1
公开(公告)日：2019-07-04
申请号：US16224644
申请日：2018-12-18
Applicant: QUALCOMM Incorporated
CPC classification number: G06K9/6202 , G06K9/00248 , G06K9/00268 , G06K9/00281 , G06K9/00288 , G06K9/00926 , G06K9/42 , G06K9/46 , G06K9/527 , G06T7/50 , G06T2207/10016
Abstract: Techniques and systems are provided for determining features for one or more objects in one or more video frames. For example, an image of an object, such as a face, can be received, and features of the object in the image can be identified. A size of the object can be determined based on the image, for example based on inter-eye distance of a face. Based on the size, either a high-resolution set of features or a low-resolution set of features is selected to compare to the features of the object. The object can be identified by matching the features of the object to matching features from the selected set of features.
-
Patent Agency Ranking
公开(公告)号：US20180268563A1
公开(公告)日：2018-09-20
申请号：US15645555
申请日：2017-07-10
Applicant: QUALCOMM Incorporated
Inventor： Ying Chen , Zhongmin Wang , Lei Wang
CPC classification number: G06T7/70 , G06K9/6202 , G06T7/90 , G06T2207/10016 , G06T2207/30196 , G06T2207/30232 , G06T2210/12
Abstract: Methods, apparatuses, and computer-readable media are provided for maintaining blob trackers for video frames. For example, a blob tracker maintained for a current video frame is identified. The blob tracker is associated with a blob detected in one or more video frames. The blob includes pixels of at least a portion of a foreground object in the one or more video frames. A current bounding region of the blob tracker for the current video frame is compared to a previous bounding region of the blob tracker for a previous video frame that is obtained earlier in time than the current video frame. It can be determined whether the current bounding region has decreased in size as compared to a size of the previous bounding region, and whether a first color characteristic of pixels of the current video frame included in the previous bounding region is within a threshold from a second color characteristic of pixels of the previous video frame included in the previous bounding region. In some examples, the blob is tracked in the current frame using the current bounding region when the current bounding region has decreased in size and when the first color characteristic is within the threshold from the second color characteristic. In some examples, the blob is tracked using the current bounding region when the blob tracker is determined to be lost (e.g., the blob tracker is not associated with the blob in the current video frame).
-
公开(公告)号：US10026193B2
公开(公告)日：2018-07-17
申请号：US15229456
申请日：2016-08-05
Applicant: QUALCOMM Incorporated
Inventor： Jinglun Gao , Ning Bi , Ying Chen , Lei Wang
Abstract: Techniques and systems are provided for processing video data. For example, techniques and systems are provided for determining costs for blob trackers and blobs. A blob can be detected in a video frame. The blob includes pixels of at least a portion of a foreground object. A physical distance between a blob tracker and the blob can be determined. A size ratio between the blob tracker and the blob can also be determined. A cost between the blob tracker and the blob can then be determined using the physical distance and the size ratio. In some cases, a spatial relationship between the blob tracker and the blob is determined, in which case the physical distance can be determined based on the spatial relationship. Blob trackers can be associated with blobs based on the determined costs between the blob trackers and the blobs.
-
公开(公告)号：US20170310972A1
公开(公告)日：2017-10-26
申请号：US15134183
申请日：2016-04-20
Applicant: QUALCOMM Incorporated
Inventor： Lei Wang , Ying Chen , Ning Bi , Xuerui Zhang , Lei Ma , Chinchuan Chiu , Jianfeng Ren
IPC: H04N19/159 , H04N19/182 , H04N19/172 , H04N19/58 , H04N19/105
CPC classification number: H04N19/159 , G06T7/254 , H04N19/105 , H04N19/172 , H04N19/182 , H04N19/23 , H04N19/58
Abstract: Techniques and systems are provided for generating a background picture. The background picture can be used for coding one or more pictures. For example, a method of generating a background picture includes generating a long-term background model for one or more pixels of a background picture. The long-term background model includes a statistical model for detecting long-term motion of the one or more pixels in a sequence of pictures. The method further includes generating a short-term background model for the one or more pixels of the background picture. The short-term background model detects short-term motion of the one or more pixels between two or more pictures. The method further includes determining a value for the one or more pixels of the background picture using the long-term background model and the short-term background model.
-
公开(公告)号：US11386633B2
公开(公告)日：2022-07-12
申请号：US17109045
申请日：2020-12-01
Applicant: QUALCOMM Incorporated
Inventor： Peng Liu , Lei Wang , Zhen Wang , Ke-Li Cheng , Ning Bi
Abstract: Systems and techniques are provided for facial image augmentation. An example method can include obtaining a first image capturing a face. Using the first image, the method can determine, using a prediction model, a UV face position map including a two-dimensional (2D) representation of a three-dimensional (3D) structure of the face. The method can generate, based on the UV face position map, a 3D model of the face. The method can generate an extended 3D model of the face by extending the 3D model to include region(s) beyond a boundary of the 3D model. The region(s) can include a forehead region, a region surrounding at least a portion of the face, and/or other region. The method can generate, based on the extended 3D model, a second image depicting the face in a rotated position relative to a position of the face in the first image.
-
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
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.
-
-
-
-
-
-
-
-
-
Search Results
39
records
(took 0.029 seconds)
