Tencent Technology (shenzhen) Company Limited

Shenzhen Innotrik Technology Co., Ltd.

Huawei Technologies Co., Limited

Zte Corporation

Chao Huang

Wei Luo

Bin Hua Liu

Runjia Huang

Ying Qi

Wanxiang Huang

Wei Li

Nan Li

Pinlin Chen

Beiqi Wu

Wenbin Zhang

Fan Ding

Like Zhang

Chunmei Deng

A model training method and apparatus, a storage medium, and a device, relating to the field of artificial intelligence (AI) technologies are provided. The method includes: obtaining a plurality of training samples, each training sample including an interaction screen and an action label, the action label indicating an interaction action adopted by a character object in the interaction screen; extracting features from the interaction screens included in the plurality of training samples, and performing clustering according to the extracted features, to obtain a clustering result; determining at least one key sample from the plurality of training samples according to the clustering result; and setting a weight for each training sample, and updating a network parameter of a deep network based on the plurality of training samples with the weights, the weight of each key sample being greater than the weight of another sample in the plurality of training samples.

Model training method and apparatus, storage medium, and device

A processing circuitry obtains a training set based on an interaction process of a plurality of cards. The training set includes a plurality of video frames, and a video frame includes a trigger region for triggering an action during the interaction process. The processing circuitry determines feature regions in the trigger region. A feature region includes a card and is set with an action label for indicating a first training content in the feature region. The processing circuitry also determines a feature vector based on the plurality of video frames. The feature vector indicates a triggering of the feature region. The processing circuit inputs the feature vector into a first model for training to obtain a second model. The first model associates the feature vector with the action label, and the second model is used for indicating a correspondence between a content of the card and the action label.

Model training method, content generation method, and related apparatuses

Speakerphone

A computer-implemented method for downloading and displaying pictures associated with instant messages is performed at a computing device. The computing device displays a list of instant messages on the screen, at least one of the instant messages having an associated picture and including address information of the picture. While the picture is being downloaded, the computing device detects a user instruction to slide the list of instant messages on the screen. In response to the user instruction, the computing device pauses the download of the picture and determines, among the list of instant messages, a second instant message having an associated second picture and including address information of the second picture based on the user instruction. The computing device then downloads the second picture associated with the second instant message according to the address information without receiving a download instruction from the user of the computing device.

Method and apparatus for downloading and displaying pictures

Method and device of sharing photos includes: receiving, from a mobile device, a first location of a first photo, a second location of a second photo, and a current location of the mobile device; sending, to the mobile device, a plurality of points of interest selected in accordance with their proximity to the first location, the second location, and the current location, respectively; receiving a first check-in location for the first photo and a second check-in location for the second photo selected from the plurality of points of interest; determining a granularity level at which an encompassing location comprises the first check-in location corresponding to the first photo and the second check-in location corresponding to the second photo; determining the encompassing location; and causing the encompassing location to be shared among users of the social network platform in association with the first photo and the second photo.

Photo check-in method, apparatus, and system

A computer-implemented method for downloading and displaying pictures associated with instant messages is performed at a computing device. The computing device receives an instant message having an associated picture and including address information of the picture. The computing device determines whether it is in a first network environment. If so, the computing device displays a thumbnail of the picture on the screen and a visual cue indicating that the picture is being downloaded and starts downloading the picture associated with the instant message according to the address information without receiving a download instruction from a user of the computing device. The thumbnail is then replaced with the picture after the picture is downloaded. While the picture is being downloaded, the computing device may pause the download when it is no longer in the first network environment and replaces the visual cue with a different one indicating so.

Method and device of sharing photos and associated information among users of a social network platform are disclosed. The method includes: a mobile device sending, to a server, a first location of a first photo and a current location of the mobile device; receiving a first plurality of POIs, at least one of which is located near the first location of the first photo and another one of which is located near the current location of the mobile device; determining a first POI among the first plurality of POIs as a first check-in location of the first photo; and sending the first photo and the first POI to the server.

Growing community of inventors

Shenzhen, China

Chao Huang