-
公开(公告)号：US20240223813A1
公开(公告)日：2024-07-04
申请号：US18392557
申请日：2023-12-21
发明人： Bolin CHEN , Jie CHEN , Shurun WANG , Yan YE , Shiqi WANG
IPC分类号： H04N19/70 , H04N19/124 , H04N19/136 , H04N19/172
CPC分类号： H04N19/70 , H04N19/124 , H04N19/136 , H04N19/172
摘要： A method of decoding a bitstream to output one or more pictures for a video stream, includes: receiving a bitstream; and decoding, using coded information of the bitstream, one or more pictures. The decoding includes: determining, based on an identifying number, whether a face video generative compression scheme is used; in response to a determination that the face video generative compression scheme is used, decoding a supplemental enhancement information (SEI) message, the SEI message comprising facial information; and reconstructing a face picture based on the facial information and a base picture associated with the SEI message.
-
公开(公告)号：US20240221363A1
公开(公告)日：2024-07-04
申请号：US18400954
申请日：2023-12-29
发明人： Binzhe Li , Shiqi Wang , Yan Ye , Shurun Wang
CPC分类号： G06V10/7715 , G06V10/273 , G06V10/34 , G06V10/806 , G06V10/82 , G06V20/46 , G06V20/49
摘要： Methods and systems implement input picture data preprocessing for a learning model by picture data blurring based on deep features. Intermediate features are extracted from convolutional layers of a preprocessing model, and each set of intermediate features are fused to yield a fused feature map, and enlarged to input picture size. Based on the fused feature map, the preprocessing model can configure one or more processors of an input preprocessing computing system to, in performing blurring preprocessing computations, emphasize picture data having larger corresponding characteristic values, and deemphasize other picture data.
-
公开(公告)号：US11924409B2
公开(公告)日：2024-03-05
申请号：US17651639
申请日：2022-02-18
发明人： Xinwei Li , Jie Chen , Ru-Ling Liao , Yan Ye
IPC分类号： H04N19/105 , H04N19/132 , H04N19/159 , H04N19/176
CPC分类号： H04N19/105 , H04N19/132 , H04N19/159 , H04N19/176
摘要： A video processing method includes: determining whether an inter predictor correction is enabled for a coding block; and when the inter predictor correction is enabled for the coding block, performing the inter predictor correction by: obtaining a plurality of predicted samples from a top boundary and a left boundary of a predicted block corresponding to the coding block; obtaining a plurality of reconstructed samples from top neighboring reconstructed samples and left neighboring reconstructed samples of the coding block; and deriving a corrected predicted block based on the plurality of the predicted samples, the plurality of the reconstructed samples and the predicted block.
-
公开(公告)号：US20240045975A1
公开(公告)日：2024-02-08
申请号：US18066207
申请日：2022-12-14
发明人： SHUANGCHEN LI , ZHE ZHANG , LINYONG HUANG , DIMIN NIU , XUANLE REN , HONGZHONG ZHENG
CPC分类号： G06F21/602 , G06F21/54
摘要： The present disclosure discloses a processor and a multi-core processor. The processor includes a processor core and a memory. The processor core includes a homomorphic encryption instruction execution module and a general-purpose instruction execution module; the homomorphic encryption instruction execution module is configured to perform homomorphic encryption operation and includes a plurality of instruction set architecture extension components, wherein the plurality of instruction set architecture extension components are respectively configured to perform a sub-operation related to the homomorphic encryption; the general-purpose instruction execution module is configured to perform non-homomorphic encryption operation. The memory is vertically stacked with the processor core and is used as a cache or scratchpad memory of the processor core.
-
公开(公告)号：US20240005133A1
公开(公告)日：2024-01-04
申请号：US17899557
申请日：2022-08-30
发明人： Linyong HUANG , Zhe ZHANG , Shuangchen LI , Hongzhong ZHENG
摘要： This application describes an hardware and a software design for quantization in GNN computation. An exemplary method may include: receiving a graph comprising a plurality of nodes respectively represented by a plurality of feature vectors; segmenting the plurality of feature vectors into a plurality of sub-vectors and grouping the plurality of sub-vectors into a plurality of groups of sub-vectors; performing vector clustering on each of the plurality of groups of sub-vectors to generate a plurality of centroids as a codebook; encoding each of the plurality of feature vectors to obtain a plurality of index maps by quantizing sub-vectors within the each feature vector based on the codebook, wherein each index map occupies a smaller storage space than the each feature vector does; and storing the plurality of index maps as an assignment table instead of the plurality of feature vectors to represent the plurality of nodes for GNN computation.
-
86.发明公开公开(公告)号：US20240005075A1
公开(公告)日：2024-01-04
申请号：US18071970
申请日：2022-11-30
发明人： Shuangchen LI , Dimin NIU , Hongzhong ZHENG
IPC分类号： G06F30/34 , G06F30/331
CPC分类号： G06F30/34 , G06F30/331
摘要： This application describes systems and methods for facilitating memory access for graph neural network (GNN) processing. An example method includes fetching, by an access engine circuitry implemented on a circuitry board, a portion of structure data of a graph from one or more of a plurality of flash memory drives implemented on the circuitry board; performing node sampling using the fetched portion of the structure data of the graph to select one or more sampled nodes; fetching a portion of attribute data of the graph from two or more of the plurality of memory drives in parallel according to the selected one or more sampled nodes; sending the fetched portion of the attribute data of the graph to a host outside of the circuitry board; and performing, by the host, GNN processing for the graph using the fetched portion of the attribute data of the graph.
-
87.发明公开公开(公告)号：US20230417892A1
公开(公告)日：2023-12-28
申请号：US18171259
申请日：2023-02-17
发明人： Pengyu Zhang , Bo Liang , Xue Wang , Hongqiang Liu
IPC分类号： G01S13/26
CPC分类号： G01S13/26
摘要： A radio frequency communication method, a vehicle control method, a device, and a system are provided. The radio frequency communication method includes: acquiring a first response signal and a second response signal which are corresponding to a radio frequency tag, wherein the first response signal and the second response signal are respectively generated by the radio frequency tag responding to radio frequency signals of different powers; determining, based on the second response signal, positioning information of the radio frequency tag; and establishing, based on the positioning information and the first response signal, a communication connection to the radio frequency tag.
-
公开(公告)号：US11856243B2
公开(公告)日：2023-12-26
申请号：US17681875
申请日：2022-02-28
发明人： Jin Xu , Ding Jiandong
IPC分类号： G06F15/16 , H04N21/2187 , H04L65/403 , H04W4/06 , H04L67/14
CPC分类号： H04N21/2187 , H04L65/403 , H04L65/4046 , H04L67/14 , H04W4/06
摘要： Provided in the example embodiments are a method, electronic device, and computer storage medium for managing virtual streaming. The method for managing virtual streaming includes: providing a configuration interface for configuring a virtual streaming room with configuration settings, the configuration settings including at least an audience group setting, and a host setting related to a virtual host; receiving, via the configuration interface, configuration input corresponding to the audience group setting, and/or the host setting; and based on the configuration input, generating a host assignment rule, the host assignment rule for assigning at least one virtual host to at least one audience group. The example embodiments solve the problems in which the form of an existing virtual host is fixed, audience adhesion is poor, and use thereof is inflexible.
-
公开(公告)号：US20230385679A1
公开(公告)日：2023-11-30
申请号：US17804664
申请日：2022-05-31
发明人： Jiachen HUANG , Dawei DING , Jianxin CHEN
摘要： Systems and methods are disclosed for benchmarking a set of quantum gates. Consistent with disclosed embodiments, a benchmarking method can include obtaining a sequence of M quantum gates from a group of quantum gates according to a probability distribution. The quantum gates in the group can be capable of polynomial-time classical simulation. The method can further include obtaining an outcome measure by applying the sequence of M quantum gates to N qubits of a quantum computing device and obtaining a probability of obtaining the outcome value given the application of the selected sequence of quantum gates. The probability can be obtained using classical simulation of the selected sequence of quantum gates. A fidelity benchmark for the M quantum gates can be generated based at least in part on the obtained probability.
-
公开(公告)号：US20230306257A1
公开(公告)日：2023-09-28
申请号：US17866194
申请日：2022-07-15
发明人： Fei SUN , Minghai QIN , Haoran LI , Guocai ZHU , Yuan GAO , Guyue HUANG , Yawen ZHANG
摘要： Neural network (NN) model training techniques can include computing activations in a forward pass using a sparse weight matrix that is transpose invariant. The neural network (NN) model training techniques can further include computing activation gradients and weight gradients in a backward pass using the sparse weight matrix.
-
-
-
-
-
-
-
-
-
搜索结果
122 条记录 (耗时 0.034 秒)
