公开(公告)号：WO2024001894A1

公开(公告)日：2024-01-04

申请号：PCT/CN2023/101533

申请日：2023-06-20

Applicant: 华为技术有限公司 ,  HUAWEI TECHNOLOGIES CO., LTD. ,  清华大学深圳国际研究生院 ,  TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL

Inventor： 侯博文 ,  HOU, Bowen ,  刘明源 ,  LIU, Mingyuan ,  巩马理 ,  GONG, Mali ,  马建设 ,  MA, Jianshe ,  邹宇豪 ,  ZOU, Yuhao

IPC: G11B7/24097

Abstract: 本申请公开了一种信息验证方法，该方法中，可以获取荧光信号，荧光信号为基于多个荧光光斑而生成的电信号，荧光信号中包含多个脉冲信号；基于预设幅度阈值以及每个脉冲信号的幅度，确定每个脉冲信号对应的信息符；根据每个脉冲信号对应的信息符，获得信息序列；根据多个脉冲信号中任意两个脉冲信号之间的脉冲信号的第一数量，以及任意两个脉冲信号之间的时间间隔，获得关于信息序列的验证结果。通过该方法，可以解决目前在光盘等技术领域，传统的信息读出方式很容易由于时序上的不稳定或抖动而导致读出信号出现偏差，从而导致读出信号对应的信息内容出现错误的问题。

公开(公告)号：WO2023088531A1

公开(公告)日：2023-05-25

申请号：PCT/EP2021/081770

申请日：2021-11-16

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  AGAPITOS, Alexandros

Inventor： MILENOVIC, Aleksandar ,  ZHOU, Mingqi

IPC: H04L41/142 ,  H04L41/16 ,  G06N3/02 ,  G06N20/00 ,  H04L41/5009 ,  H04L43/08

Abstract: The present disclosure relates to a Network Element (NE) and a management entity for supporting Anomaly Detection (AD) in a communication network. The NE is configured to determine, based on a local AD model, a point to be an anomalous point in Key Performance Indicator (KPI) time-series data, provide the anomalous point to the management entity, receive a label for the points of the KPI time series data from the management entity, and update the local AD model based on the received label. The management entity is configured to receive, from the NE, a point determined to be an anomalous point, and provide to the NEs, a label for the point. The management entity is further configured to receive, from the NE, a parameter of an updated local AD model of that NE, and update a global AD model based on the received parameter.

公开(公告)号：WO2023087538A1

公开(公告)日：2023-05-25

申请号：PCT/CN2022/073253

申请日：2022-01-21

Applicant: HUAWEI TECHNOLOGIES CO.,LTD.

Inventor： ASHWOOD-SMITH, Peter ,  AKHAVAIN MOHAMMADI, Mehdi Arashmid

IPC: H04B7/185

Abstract: For a communication network using a satellite-involved backhaul, the backhaul outage and restoration states are predicted based on satellite motion data. Based on such predictions, devices providing the core portion of the communication network, and nearby Internet or backhaul radio devices can schedule or take actions. Actions can include but are not necessarily limited to: powering equipment up or down, suspending communications, migrating software from servers being powered down, transmitting replies to packets to indicate an anticipated outage and optionally anticipated outage end time, marking packets with congestion indications, closing or reopening certain ports, withdrawing or reinstating routing table addresses, and transmitting outage notifications to users or devices.

公开(公告)号：WO2023085968A1

公开(公告)日：2023-05-19

申请号：PCT/RU2021/000505

申请日：2021-11-15

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  SOLODSKIKH, Kirill Igorevich

Inventor： KURBANOV, Azim Edgarovich ,  AYDARKHANOV, Ruslan Daurenovich ,  SONG, Dehua ,  FILIPPOV, Alexander Nikolaevich ,  YAN, Youliang

IPC: G06N3/04 ,  G06N3/08 ,  G06N7/00

Abstract: The present disclosure relates to methods and devices for performing neural network pruning in the field of machine learning. In particular, each layer of a trained neural network that comprises a first set of discrete weights is represented by a continuous parameter representation. The continuous parameter representation is based on a linear combination of Riemann integrable functions. Then, said continuous parameter representation may be discretized to obtain a second set of discrete weights of a desired size for that layer. In this way, neural network pruning is performed and the size of the neural network may be changed at the inference phase. Moreover, there is no need to fine-tune the second set of discrete weights because of the continuous parameter representation.

公开(公告)号：WO2023085962A1

公开(公告)日：2023-05-19

申请号：PCT/RU2021/000496

申请日：2021-11-11

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  KARABUTOV, Alexander Alexandrovich

Inventor： JIA, Panqi ,  BOEV, Atanas ,  GAO, Han ,  WANG, Biao ,  SAUER, Johannes ,  ALSHINA, Elena Alexandrovna

IPC: H04N19/186 ,  H04N19/12 ,  H04N19/436 ,  H04N19/91

Abstract: The present disclosure relates to conditional coding of components of an image. It is provided a method of encoding at least a portion of an image, comprising encoding a primary component of the image independently from at least one secondary component and encoding the at least one secondary component of the image using information from the primary component. Further, it is provided a method of encoding at least a portion of an image, comprising providing a residual comprising a primary residual component for a primary component of the image and at least one secondary residual component for at least one secondary component of the image that is different from the primary component, encoding the primary residual component independently from the at least one secondary residual component and encoding the at least one secondary residual component using information from the primary residual component.

公开(公告)号：WO2023083476A1

公开(公告)日：2023-05-19

申请号：PCT/EP2021/081698

申请日：2021-11-15

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  TANAY, Thomas

Inventor： LEONARDIS, Ales ,  MAGGIONI, Matteo

IPC: G06T5/00 ,  G06K9/62 ,  G06T5/50

Abstract: The present disclosure provides an image processing apparatus for mitigating the effects of image degradation, the image processing apparatus comprising one or more processors configured to: receive two or more images depicting at least a portion of a common subject; calculate at least one optical offset between the two or more images;extract, from each of the two or more images, one or more feature representations, each of which represents tensor data based on the two or more images; extract, from the one or more feature representations, a plurality of patches, which represent one or more characteristics of the two or more images; generate, for each of the plurality of patches, a warp corrected patch by applying a warping operation, the warping operation being based on the calculated at least one optical offset between the two or more images; generate, for each of the plurality of patches, a plurality of kernels based on each warp corrected patch; correct, for each of the warp corrected patches, distortions of the characteristics of the two or more images based on the plurality of kernels; generate, for each of the warp corrected patches, a convolved patch by applying a convolution; and combine each of the convolved patches into an output feature representation. This provides the ability to account for motion in the scene when combining images, adaptively process and aggregate corresponding regions in different image frames and maximize restoration quality at the end of the image processing pipeline.

公开(公告)号：WO2023083467A1

公开(公告)日：2023-05-19

申请号：PCT/EP2021/081616

申请日：2021-11-15

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  UNIVERSITY OF ZURICH

Inventor： TULYAKOV, Stepan ,  BOCCHICCHIO, Alfredo ,  GEORGOULIS, Stamatios ,  LI, Youanyou ,  GEHRIG, Daniel ,  SCARAMUZZA, Davide ,  GEHRIG, Mathias

IPC: G06T7/246 ,  G06V10/82 ,  H04N7/01

Abstract: An image processing apparatus for generating an interpolated frame at a specified time between adjacent image frames of a video. The image processing apparatus includes an input module configured to receive one, two or more key image frames, and a plurality of surrounding events. The image processing apparatus includes a parametric motion model estimator configured to estimate an inter-frame motion based on the key image frames and the plurality of surrounding events, and a warping encoder configured to compute a plurality of multiscale warping interpolation features. The image processing apparatus includes a synthesis encoder configured to compute a plurality of multiscale synthesis interpolation features, and a multiscale feature fusion decoder module configured to generate the interpolated frame. The image processing apparatus enables highly efficient and accurate video frame interpolation in slow-speed motion as well as in high-speed motion.

公开(公告)号：WO2023083466A1

公开(公告)日：2023-05-19

申请号：PCT/EP2021/081609

申请日：2021-11-15

Applicant: HUAWEI TECHNOLOGIES CO., LTD. ,  UNIVERSITY OF ZURICH

Inventor： GEORGOULIS, Stamatios ,  MESSIKOMMER, Nico ,  TULYAKOV, Stepan ,  ERBACH, Julius ,  BOCCHICCHIO, Alfredo ,  GEHRIG, Daniel ,  LI, Yuanyou ,  SCARAMUZZA, Davide

IPC: G06T5/50

Abstract: A HDR imaging device comprising: an image sensor configured to capture a plurality of LDR images; an aligned event sensor with pixels configured to capture a plurality of events between the plurality of LDR images; an image encoder module configured to generate an image feature representation based on each of the LDR images; an event encoder module configured to generate an event feature representation based on a subset of the plurality of events. A feature alignment module configured to: estimate per-pixel offsets between the image feature representation of a specified LDR image and each other LDR image and generate a plurality of aligned feature representations by applying deformable convolutions or warping to the image feature representations based on the estimated offsets. A fusion module configured to generate fused feature representation; and an image decoder module configured to generate a HDR image based on the fused feature representation.

公开(公告)号：WO2023083232A1

公开(公告)日：2023-05-19

申请号：PCT/CN2022/131001

申请日：2022-11-10

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： MEHRVAR, Hamid

IPC: H04L12/46

Abstract: A method to select a number of fibers for ROADM-equipped nodes of an optical network by which a controller is operative to determine which links are utilized as well as their usage frequencies and then partition a scale of usage frequencies into a number of intervals. By assigning a number of fibers to each one of the intervals, a number of fibers is assigned to each link, according to their usage frequencies, setting the degree for ROADMs at the nodes. Simulations can evaluate the network's performance in terms of a blocking rate representing an overall signal blocking rate by the ROADMs at network nodes. The number of intervals, their ranges, and the number of fibers associated with each interval can be iterated until an improved or satisfactory network performance is achieved.

公开(公告)号：WO2023082727A1

公开(公告)日：2023-05-19

申请号：PCT/CN2022/109837

申请日：2022-08-03

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： SINGH, Gaganpreet ,  ZHOU, Wei ,  IRANI, Pourang Polad ,  LI, Wei

IPC: G06F3/01

Abstract: Method and systems for controlling a display device, including detecting a mid-air gesture using a sensing device; mapping the detected mid-air gesture to locations of an interaction region, the interaction region including an on-screen region of the display device and an off-screen region that is located outside an edge of the on-screen region; and performing a display device control action upon detecting an edge interaction based on the mapping of the detected mid-air gesture to locations that interact with the edge of the on-screen region.