Patents Assigned to BIGO TECHNOLOGY PTE. LTD.
-
Patent number: 12784419Abstract: Provided is a method for training a scene recognition model. The scene recognition model includes a core feature extraction layer, a global information feature extraction layer, an LCS module of at least one level with an attention mechanism, and a fully-connected decision layer. The method includes: acquiring parameters of the core feature extraction layer and the global information feature extraction layer by training based on a first scene label of a sample image and a standard cross-entropy loss; training a weight parameter of the LCS module of each level, based on a loss value acquired by performing a pixel-by-pixel calculation on a feature map output from the LCS module of each level and the first scene label of the sample image; and acquiring a parameter of the fully-connected decision layer by training based on the first scene label of the sample image and the standard cross-entropy loss.Type: GrantFiled: September 30, 2022Date of Patent: November 3, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Xiongwen Luo, Jianghu Lu, Wei Xiang
-
Patent number: 12749337Abstract: Provided is a face tracking method. The method includes: in a process of tracking a face in a video frame, determining whether an optimization thread is running; in response to the optimization thread running and the video frame being a key frame, updating a second keyframe data set based on the video frame; in response to receiving a clear instruction from the optimization thread, clearing the video frame in the second keyframe data set, and updating the second keyframe data set to a first keyframe data set; in response to the optimization thread not running and the video frame being the key frame, updating the first keyframe data set based on the video frame and the second keyframe data set; and making the optimization thread optimize a facial identity based on the first keyframe data set by invoking the optimization thread upon updating the first keyframe data set.Type: GrantFiled: January 4, 2022Date of Patent: September 29, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Wenyu Chen, Gengdai Liu
-
Patent number: 12732608Abstract: Provided is a video coding processing method. The method includes: partitioning a to-be-coded unit into a plurality of to-be-coded sub-blocks; determining residual proportion information of the to-be-coded unit based on residual information of each of the plurality of to-be-coded sub-blocks; determining whether to perform an intra sub-partition check on the to-be-coded unit based on the residual proportion information; and in response to a determination result of not performing the intra sub-partition check on the to-be-coded unit, performing coding processing on the to-be-coded unit based on an adjacent coded unit.Type: GrantFiled: June 30, 2023Date of Patent: September 8, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Yihang Li, Guisen Xu
-
Patent number: 12707024Abstract: Provided is a method for processing video calls. The method includes: determining a layer bit rate corresponding to each frame layer in video data sent by a transmitter, wherein a plurality of video frames in the video data are organized into a plurality of frame layers based on an inter-frame dependency relationship, and the layer bit rate comprises a bit rate acquired by counting bit rates of video frames in a corresponding frame layer; determining, based on uplink bit rates corresponding to a plurality of receivers and a total downlink detected bandwidth, downlink allocated bandwidths corresponding to the receivers; determining, based on the downlink allocated bandwidths corresponding to the receivers and the layer bit rates, target frame layers corresponding to the receivers; and forwarding video data subjected to frame extraction processing for the target frame layers to corresponding receivers.Type: GrantFiled: August 17, 2022Date of Patent: August 11, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Yunchang Xia, Yong Deng
-
Patent number: 12695892Abstract: Provided is a method for updating a code table. The method includes: acquiring source stream video data; inputting the source stream video data into a trained preference perception model, wherein the preference perception model analyzes the source stream video data and outputs a video quality distribution table, the video quality distribution table recording prediction preference probabilities corresponding to combinations of different second video code rates and second video resolutions therein, wherein the prediction preference probabilities are predicted values of preference probabilities of a user for inputted video data at combinations of different video code rates and video resolutions; and updating the code table based on the video quality distribution table, wherein the code table records combinations of first video code rates and first video resolutions at different video definition levels.Type: GrantFiled: July 12, 2023Date of Patent: July 28, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Shuai Wang
-
Patent number: 12689776Abstract: Provided is a method for preloading a live stream in a video stream. The method includes: acquiring current device environment information and current service environment information in response to detecting a preload trigger scenario being satisfied during play of video stream data, wherein the preload trigger scenario comprises: current play data being a short video and next play data being a live video in a case of switching play content by a swipe-up operation; or current play data being a short video and previous play data being a live video in a case of switching play content by a swipe-down operation; determining whether the device environment information meets a preload device condition, and determining whether the service environment information meets a preload service condition; and preloading a live stream resource in response to determination results that the preload device condition and the preload service condition are met.Type: GrantFiled: March 13, 2023Date of Patent: July 21, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Ruijia Su
-
Patent number: 12683957Abstract: Provided is a method for authenticating identity.Type: GrantFiled: February 25, 2022Date of Patent: July 14, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Yan Li
-
Recommendation method and apparatus, and method and apparatus for publishing online prediction model
Patent number: 12670360Abstract: Provided is a recommendation method. The method includes: determining a target user identifier and a plurality of candidate item identifiers; acquiring the intermediate user feature vector corresponding to the target user identifier by searching for the target user identifier; acquiring the intermediate item feature vectors corresponding to the plurality of candidate item identifiers by searching for the plurality of candidate item identifiers; outputting degrees of matching of the plurality of candidate item identifiers and the target user identifier by predicting the intermediate user feature vector and the intermediate item feature vectors in the online prediction model; and determining a target item identifier to be recommended to the target user identifier from the plurality of candidate item identifiers.Type: GrantFiled: June 1, 2021Date of Patent: June 30, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Wanpeng Yang -
Patent number: 12633055Abstract: Provided is a method for training parameter estimation models. The method includes: estimating reconstruction parameters specified for three-dimensional face reconstruction by inputting training samples in a face image training set to a pre-constructed neural network model, and reconstructing three-dimensional faces corresponding to the training samples by inputting the reconstruction parameters to a pre-constructed three-dimensional morphable model; calculating a plurality of loss functions of a plurality of pieces of two-dimensional supervision information between the three-dimensional faces and the training samples, and adjusting weights corresponding to the plurality of loss functions; and generating fitting loss functions based on the plurality of loss functions and the weights corresponding to the plurality of loss functions, and acquiring a trained parameter estimation model by performing an inverse correction on the neural network model using the fitting loss functions.Type: GrantFiled: October 22, 2021Date of Patent: May 19, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Xiaowei Zhang, Zhongyuan Hu, Gengdai Liu
-
Patent number: 12632979Abstract: A method for training models is provided. The method includes: inputting an image training sample corresponding to a current iteration into a current posture detection network model, and acquiring a first loss function corresponding to the current iteration; re-projecting the current output result of the current posture detection network model, and acquiring a second loss function corresponding to the current iteration; and acquiring a posture detection network model for a next iteration by performing backpropagation on the current posture detection network model, and achieving training of the posture detection network model by performing the next iteration before an iteration end condition is met.Type: GrantFiled: October 26, 2021Date of Patent: May 19, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Sen Jia
-
Patent number: 12614028Abstract: Provided is a language labeling method, including: determining a language classifier; collecting a plurality of pieces of information correlated to video data, and determining the plurality of pieces of information as a plurality of pieces of video information; dividing the plurality of pieces of video information into target information and reference information; identifying languages to which the plurality of pieces of video information is attributed by inputting the plurality of pieces of video information into the language classifier; and checking a confidence level of a target language with an aid of reference languages, wherein the target language is a language to which the target information is attributed and the reference languages are a plurality of languages to which the reference information is attributed.Type: GrantFiled: December 28, 2021Date of Patent: April 28, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Zhiqiang Tu
-
Patent number: 12608547Abstract: Provided is a method and a device for segmenting words, and a storage medium. The method includes: acquiring a plurality of groups of high-resource language (HRL) data, and acquiring a plurality of groups of first word segmentation language data by processing the plurality of groups of HRL data; acquiring a plurality of groups of low-resource language (LRL) data, acquiring a plurality of candidate word segments, and selecting second word segmentation language data from the plurality of candidate word segments; acquiring a word segmentation model by training based on the second word segmentation language data, and outputting a plurality of candidate word segmentation results; and selecting the candidate word segmentation result with a highest matching degree as a word segmentation result based on a matching degree between each of the candidate word segmentation results and the first word segmentation corpus.Type: GrantFiled: January 10, 2022Date of Patent: April 21, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Jianning Zhang
-
Patent number: 12585815Abstract: Provided are an information management system and method for a communication application, and a display terminal. The system includes a personal information display module and a posted information display module; the personal information display module is configured to display personal information of a poster; and the posted information display module is configured to dynamically display a plurality of posted contents contained in a moment posted by the poster at current time.Type: GrantFiled: August 7, 2020Date of Patent: March 24, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Rong Huang, Yanshan Feng, Tiejun Wang
-
Patent number: 12581057Abstract: Provided is a method for video predictive coding. The method includes: acquiring decision information related to a current coding block in inter-frame predictive coding, wherein the decision information includes one of: pre-analysis information determined by a coder in a lightweight video coding pre-analysis, coding information of a plurality of sub-blocks acquired by recursively coding the current coding block, and information of an executed mode determined based on the executed mode; and determining, based on the decision information, whether to skip motion estimation (ME) coding of the current coding block.Type: GrantFiled: December 3, 2021Date of Patent: March 17, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Tongbing Cui
-
Patent number: 12526461Abstract: Provided is a method for allocating resource objects based on live streaming rooms. The method is applicable to a primary node on a server side, and includes: acquiring synchronization data and local data corresponding to a predetermined resource object in a target live streaming room, wherein the synchronization data is synchronously stored between the primary node and a secondary node, and includes a waiting queue of waiting users; and the local data is stored in the primary node, and includes second user data of an invited user; determining a target user based on a sequence of the waiting users in the waiting queue; acquiring a current state of the target user by comparing the synchronization data and the local data for the target user; and determining, based on the current state, whether to send an invitation for allocating the predetermined resource object to a target user side.Type: GrantFiled: November 30, 2022Date of Patent: January 13, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Yuhui Zhi
-
Patent number: 12526129Abstract: Provided is a data encryption method. The data encryption method includes: determining source data requested to be encrypted by an application in an untrusted execution environment; generating reference data corresponding to the source data in the untrusted execution environment, wherein the reference data is data for one-time use in encrypting the source data; reading a primary key generated for the application in a trusted execution environment; acquiring security data by encrypting the reference data using the primary key in the trusted execution environment; and acquiring target data by encrypting the source data using the security data in the untrusted execution environment.Type: GrantFiled: July 28, 2021Date of Patent: January 13, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Yan Li
-
Patent number: 12518357Abstract: Provided is a video denoising method, applicable to a mobile terminal. The method includes acquiring video data, acquiring environmental parameters related to denoising in an environment of the mobile terminal, calculating an extent of conflict between the environment and the denoising based on the environmental parameters, and determining, based on the extent of conflict, a state of denoising the video data.Type: GrantFiled: April 2, 2021Date of Patent: January 6, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Min Yang, Bingyi Song
-
Patent number: 12511758Abstract: A method for detecting and tracking a target is provided. The method includes: acquiring a target tracking result of each video frame of a plurality of video frames received from a video stream by inputting the each video frame of the plurality of video frames into a tracking network; in response to a video frame of the plurality of video frames being a key frame, acquiring a target detection result output by a detection network upon receiving a last delay frame in a specified number of delay frames by inputting the key frame into the detection network and controlling the detection network to operate during a period of receiving the specified number of delay frames; and generating a final target tracking result based on the target detection result and a target tracking result of the last delay frame.Type: GrantFiled: August 6, 2021Date of Patent: December 30, 2025Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Wei Xiang, Jundong Wang, Lvran Chen, Jiening Jiao
-
Patent number: 12513365Abstract: Provided is a method for training a content moderation mode. The method includes extracting part of image data of a sample video file as sample image data; positioning a time point of the sample image data in the sample video file in the case that the sample image data contains offensive content; extracting salient image region data from the image data around the time point; and training the content moderation model based on the image region data and the sample image data.Type: GrantFiled: August 6, 2020Date of Patent: December 30, 2025Assignee: BIGO TECHNOLOGY PTE. LTD.Inventors: Feng Shi, Zhenqiang Liu
-
Patent number: D1144406Type: GrantFiled: March 31, 2025Date of Patent: August 25, 2026Assignee: BIGO TECHNOLOGY PTE. LTD.Inventor: Anrong Tan