-
公开(公告)号：US20220164571A1
公开(公告)日：2022-05-26
申请号：US17355893
申请日：2021-06-23
Applicant: BOE Technology Group Co., Ltd.
Inventor： Xiaojun TANG
Abstract: The present application provides a model training method and apparatus, an electronic device and a readable storage medium. The model training method of a pedestrian re-identification network, includes: determining a current weight matrix for multiple identification objects according to a current training result of a neural network model; calculating a difficulty vector corresponding to the multiple identification objects according to the current weight matrix; and, selecting next training samples of the multiple identification objects for a next training according to the difficulty vector.
-
公开(公告)号：US20210166042A1
公开(公告)日：2021-06-03
申请号：US16066799
申请日：2017-10-24
Applicant: BOE TECHNOLOGY GROUP CO., LTD.
Inventor： Haijun SU , Xiaojun TANG
Abstract: The disclosure provides an objective identification device, comprising: a classifier training circuit configured to extract objective characteristics based on training samples and perform offline training based on the objective characteristics to obtain a classifier; and a calculation circuit is configured to identify an objective in an image based on a particle swarm optimization algorithm, wherein each of particles is defined as an object having a predefined size in the image; and a fitness value of each of particles is calculated based on the classifier and the objective characteristics of the particle in the particle swarm optimization algorithm, the fitness value representing a probability that the particle belongs to the objective. The disclosure also provides an objective identification method and driving assistance device. According to the disclosure, not only the identification rate can be increased but also application scenarios having different identification rate requirements can be satisfied.
-
公开(公告)号：US20210019430A1
公开(公告)日：2021-01-21
申请号：US17042421
申请日：2020-01-22
Applicant: BOE TECHNOLOGY GROUP CO., LTD.
Inventor： Xiaojun TANG
Abstract: A digital artwork display device, a digital artwork management method, and an electronic device are provided. The digital artwork display device includes a registration circuit, a transaction circuit, and a file decryption circuit. The registration circuit is configured to apply for a device identifier and a device public-private key pair, and the device public-private key pair includes a device public key and a device private key. The transaction circuit is configured to acquire a use license, and the use license includes the device identifier and a content key ciphertext obtained by encrypting a content key by using the device public key. The file decryption circuit is configured to decrypt the content key ciphertext in the use license by using the device private key to obtain the content key, and decrypt an encrypted file by using the content key to obtain an original file.
-
公开(公告)号：US20190065891A1
公开(公告)日：2019-02-28
申请号：US15941941
申请日：2018-03-30
Applicant: BOE TECHNOLOGY GROUP CO., LTD.
Inventor： Jinglin YANG , Xiaojun TANG
Abstract: Embodiments of the present disclosure disclose an image detection method including: establishing a data model based on mark point information of a group of first strip-like line segments, the mark point information of each first strip-like line segment including the length, the width, the direction and the coordinate of the center point of the first strip-like line segment, establishing a priori model based on the distribution and the number of a group of second strip-like line segments which constitute a target structure, establishing a probability density function for the data model and the priori model, and performing sampling and solution optimization to obtain a globally optimal solution, and detecting the image taking the globally optimal solution as a target model. The embodiments of the present disclosure also disclose an image detection system and a non-volatile computer readable medium.
-
-
-
Search Results
24
records
(took 0.018 seconds)
