Bigo Technology Pte. Ltd.

Comba Telecom Systems (china) Limited

Wei Xiang

Qiushi Huang

Yiyong Li

Zhun Sun

Yifeng Wang

Xue Jing

Bo Yang

Zhujin Liang

Chao Pei

Shaohu Fang

Yule Li

An Li

Jiahong Huang

Provided is a method for training an image generation model, including: acquiring a first transformation model by training; acquiring a reconstruction model by training based on the first transformation model; acquiring a second transformation model by training; generating a grafted transformation model by grafting the first transformation model with the second transformation model; and generating the image generation model based on the reconstruction model and the grafted transformation model. The first transformation model is configured to generate a first training image according to a first noise sample. The first training image is an image of a first style. The reconstruction model is configured to map an original image sample to a latent variable corresponding to the original image sample. The second transformation model is configured to generate a second training image according to a second noise sample. The second training image is an image of a second style.

Method for training image generation model, method for generating images, and devices thereof

Provided is a method for detecting hand key points. The method includes: acquiring a hand image to be detected; acquiring heat maps of the hand key points by inputting the hand image into a pre-trained heat map model, wherein the heat maps include two-dimensional coordinates of the hand key points; acquiring hand structured connection information by inputting the heat maps and the hand image into a pre-trained three-dimensional information prediction model; and determining, based on the hand structured connection information and the two-dimensional coordinates in the heat maps, three-dimensional coordinates of the hand key points in a world coordinate system.

Method for detecting hand key points, method for recognizing gesture, and related devices

Provided is a method for moderating abnormal users. The method includes: acquiring history behavior data of a plurality of to-be-moderated users, wherein the history behavior data includes behavior data of interactions of the plurality of to-be-moderated users; extracting a plurality of predetermined valid features from the history behavior data for each of the plurality of to-be-moderated users, wherein the predetermined valid features are features predetermined in sample data; calculating, based on predetermined probabilities of events associated with the plurality of valid features, a probability that each of the plurality of to-be-moderated users is the abnormal user; establishing a total probability function based on probabilities that the plurality of to-be-moderated users are the abnormal users; determining candidate users by solving, based on predetermined conditions, a maximum value of the total probability function; and determining the abnormal users by moderating the candidate users.

Method and apparatus for moderating abnormal users, electronic device, and storage medium

A method for training a neural network, including: determining a neural network; training the neural network at a first learning rate according to a first optimization mode, where the first learning rate is updated each time the neural network is trained; mapping the first learning rate of the first optimization mode to a second learning rate of a second optimization mode in the same vector space; determining the second learning rate satisfies a preset update condition; and continuing to train the neural network at the second learning rate according to the second optimization mode.

Method and apparatus for detecting face, computer device and computer-readable storage medium

Provided is a method for moderating live streaming. The method includes: performing preliminary violation moderation on a current live streaming frame in a to-be-moderated live streaming room based on a high-accuracy moderation model and a high-recall moderation model that are cascaded with each other; and acquiring a target violation score of the current live streaming frame by inputting a first violation score of the current live streaming frame under the high-accuracy moderation model, a second violation score of the current live streaming frame under the high-recall moderation model, and a multi-dimensional behavior feature of the current live streaming frame in the to-be-moderated live streaming room to a pre-constructed behavior moderation model in response to the current live streaming frame passing the preliminary violation moderation.

Live streaming moderating method and apparatus, server, and storage medium

Provided is a human posture detection method including: acquiring a plurality of frames of image data; acquiring a plurality of human posture reference maps output by a human posture detection model responsive to inputting a current frame of image data to the human posture detection model with reference to human posture confidence maps of a previous frame of image data, wherein different human posture reference maps correspond different human-posture key points; identifying a human-posture key point in each of the human posture reference maps; and generating human posture confidence maps of the current frame of image data based on credibility of the human-posture key points, wherein the human posture confidence maps of the current frame of image data is configured to participate in generation of human posture confidence maps of a next frame of image data.

Human posture detection utilizing posture reference maps

The invention discloses a small cell base station system, and related devices and data processing methods, and the small cell base station system includes an extended small cell base station and at least one distributed component, where the extended small cell base station includes a network protocol processing module, an upper layer protocol processing module, a physical layer protocol processing module, a first baseband data processing module, and a first transceiver module; and the distributed component includes a second baseband data processing module and a second transceiver module, where in the downlink, the first baseband data processing module separates downlink baseband data into at least two branches, performs a data alignment operation on the at least two branches of downlink baseband data, sends one of the branches to the first transceiver module, and Ethernet encapsulates and then sends the remaining branches to the second transceiver module through the second baseband data processing module of the distributed component, thus extending a coverage area of the small cell base station system.

Growing community of inventors

Guangzhou, China

Wei Xiang