公开(公告)号：WO2022199078A1

公开(公告)日：2022-09-29

申请号：PCT/CN2021/131766

申请日：2021-11-19

Applicant: 苏宁易购集团股份有限公司

Inventor： 孙岩 ,  冀怀远 ,  蔡忠强 ,  李永翔 ,  刘澍 ,  荆伟

IPC: G06Q30/06 ,  G06K9/62 ,  G06K9/6215 ,  G06K9/6256 ,  G06K9/6277 ,  G06N3/0454 ,  G06N3/08 ,  G06Q30/0601 ,  G06Q30/0643

Abstract: 本发明公开了一种商品动态信息生成方法、装置、设备及计算机可读介质，属于计算机技术领域。方法包括：采集货架的感应信息以及顾客的动态图像；在动态图像中筛选出与感应信息对应的有效动态图像；确定与感应信息对应的第一商品信息；确定与有效动态图像中商品对应的第二商品信息；将第一商品信息与第二商品信息对比，获得商品对比结果；将商品对比结果与信息生成条件对比，若满足信息生成条件则生成商品动态信息。本发明利用与货架感应信息对应的有效动态图像获得第二商品信息，通过第二商品信息与根据感应信息获得第一商品信息对比，根据对比结果生成商品动态信息，避免了现有技术中仅根据第一商品信息生成商品动态信息而导致的识别错误的情况。

公开(公告)号：WO2022028928A1

公开(公告)日：2022-02-10

申请号：PCT/EP2021/070743

申请日：2021-07-23

Applicant: ANALOG DEVICES INTERNATIONAL UNLIMITED COMPANY

Inventor： SPALDING, George ,  HURWITZ, Jonathan ,  HOLLAND, William

IPC: H04L9/32 ,  H04L9/08 ,  G06F11/3055 ,  G06F21/72 ,  G06F21/73 ,  G06K9/6215 ,  H04L2209/34 ,  H04L9/0866 ,  H04L9/3278

Abstract: Physical Unclonable Functions, PUFs, are hardware devices designed to generate a number that is random (i.e., two identical PUFs should produce randomly different numbers from each other) and persistent (i.e., a PUF should consistently generate the same number over time). Over time, aspects of the PUF hardware may change or drift, which may ultimately cause the generated number to change, and therefore no longer be persistent. Failure to generate a persistent number may cause difficulties for other devices that rely on the persistence of the number generated by the PUF, for example as part of a cryptographic process. The present disclosure relates to monitoring over time the physical characteristics of the PUF that are used to generate its number, and thereby keep track of its reliability to generate a random number that is persistent. By monitoring PUFs in this way, it may be possible to detect in advance a PUF that is at risk of generating a number that is no longer persistent, so that pre-emptive action may be taken before the PUF actually fails.

公开(公告)号：WO2022011754A1

公开(公告)日：2022-01-20

申请号：PCT/CN2020/106703

申请日：2020-08-04

Applicant: 苏州大学

Inventor： 沈长青 ,  雷飘 ,  谢靖 ,  张爱文 ,  江星星 ,  王俊 ,  石娟娟 ,  黄伟国 ,  朱忠奎

IPC: G01M13/045 ,  G06K9/62 ,  G06F17/16 ,  G06K9/6215 ,  G06K9/6269 ,  G06K9/6276 ,  G06N20/10

Abstract: 一种基于自适应流形嵌入动态分布对齐的故障诊断方法，通过自动计算最优的子空间维数，并计算测地线流式核和变换后的流形特征表示，避免数据在原始欧式空间的特征扭曲。引入相似度度量A-distance定义一个自适应因子，动态调整样本数据条件分布和边缘分布的相对权重，缩小了源域和目标域样本的分布差异，提高了变工况下滚动轴承故障诊断的准确性和有效性。该方法可解释性强，对计算机硬件资源的要求较低，执行速度更快，同时具备出色的诊断精确度、算法收敛性和参数鲁棒性。尤其适用于变工况下多场景、多故障的轴承故障诊断，可广泛地应用于机械、电力、化工、航空等复杂系统的多变工况下的故障诊断任务。

公开(公告)号：WO2022005303A1

公开(公告)日：2022-01-06

申请号：PCT/NZ2020/050162

申请日：2020-12-01

Applicant: WETA DIGITAL LIMITED ,  GOURMEL, Olivier

Inventor： GOURMEL, Olivier

IPC: G06T13/40 ,  G06K9/00 ,  G06F17/18 ,  G06K9/6215 ,  G06K9/622 ,  G06K9/6224 ,  G06T19/00

Abstract: A computer-implemented method for processing a set of virtual fibers into a set of clusters of virtual fibers, usable for manipulation on a cluster basis in a computer graphics generation system, may include determining aspects for virtual fibers in the set of virtual fibers, determining similarity scores between the virtual fibers based on their aspects, and determining an initial cluster comprising the virtual fibers of the set of virtual fibers. The method may further include instantiating a cluster list in at least one memory, adding the initial cluster to the cluster list, partitioning the initial cluster into a first subsequent cluster and a second subsequent cluster based on similarity scores among fibers in the initial cluster, adding the first subsequent cluster and the second subsequent cluster to the cluster list, and testing whether a number of clusters in the cluster list is below a predetermined threshold.

公开(公告)号：WO2022005158A1

公开(公告)日：2022-01-06

申请号：PCT/KR2021/008165

申请日：2021-06-29

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： MODI, Rajat ,  BATHULA, Sreekar ,  NELLURU, Vishnu Teja ,  SHARMA, Manish

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/6215 ,  G06K9/6221 ,  G06K9/6256 ,  G06K9/6263 ,  G06N3/0454 ,  G06N3/088 ,  G06Q30/0201 ,  G06Q30/0623 ,  G06Q30/0643 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T7/70

Abstract: The present subject matter is related to a content-generation method in a computing-environment based on artificial neural network (ANN) such as generative adversarial network (GAN). An external-input for generation of content may be received by a Generative adversarial network (GAN); the GAN may be configured to operate in respect of a plurality of target domain attributes (TDA) for a target domain. A plurality of target domain attribute (TDA) may shortlisted from the plurality of TDA based at least one of the external input and one or more clusters associated with the plurality of TDA. Data may be interpolated within a latent space defined by representations of the shortlisted TDA.

公开(公告)号：WO2021191887A1

公开(公告)日：2021-09-30

申请号：PCT/IL2021/050242

申请日：2021-03-04

Applicant: ORBOTECH LTD.

Inventor： RAVEH, Gonen

IPC: G06K9/62 ,  G06N20/00 ,  G06K9/6215 ,  G06K9/6267

Abstract: A method, a computerized apparatus and a computer program product for adaptive learning for image classification. The method comprises applying a set of classification models on a calibration dataset and on a production dataset, and calculating disagreement measurements over the predictions thereof on each dataset. Based on similarity measurement between the disagreement measurement of the calibration and the production datasets, being below a predetermined threshold, a data drift is indicated in the production dataset. The method further comprises determining a training dataset for training a classification model for the production dataset. The training dataset is selected over a plurality of sets of images ordered according to time intervals in which images therein are obtained. The selection is performed based on weights determined for the plurality of sets.

公开(公告)号：WO2021150939A1

公开(公告)日：2021-07-29

申请号：PCT/US2021/014688

申请日：2021-01-22

Applicant: HOME DEPOT INTERNATIONAL, INC.

Inventor： AFSHAR, Estelle ,  HAGEN, Matthew ,  QU, Huiming

IPC: G06N3/04 ,  G06F16/58 ,  G06K9/46 ,  G06N3/02 ,  G06N20/00 ,  G06K9/6202 ,  G06K9/6215 ,  G06N3/0454 ,  G06N3/08 ,  G06Q30/0623

Abstract: A computer-implemented method for determining image similarity includes determining, by a first neural network, a first feature value associated with a first characteristic of a first product based on an image of the first product. The method also includes determining, by a second neural network, a second feature value associated with a second characteristic of the first product based on the image of the first product. The method further involves calculating a first vector space distance between the first feature value and a third feature value associated with the first characteristic of a second product, and calculating a second vector space distance between the second feature value and a fourth feature value associated with the second characteristic of the second product. Additionally, the method includes determining a similarity value based on the first vector space distance and the second vector space distance.

公开(公告)号：WO2021147292A1

公开(公告)日：2021-07-29

申请号：PCT/CN2020/105668

申请日：2020-07-30

Applicant: 南京航空航天大学

Inventor： 陆蔚华 ,  孙天琪

IPC: G06Q10/06 ,  A61B5/369 ,  A61B5/398 ,  G06K9/00536 ,  G06K9/6215 ,  G06K9/6292 ,  G06Q10/0639

Abstract: 一种联合脑电和眼动并结合用户相似度的产品设计方案决策方法，包含三个并行的步骤：(1)采集原始脑电信号并去除伪迹和噪声，提取EEG关于情绪唤醒度的β波段和关于情绪效价的α波段，并提取事件相关电位ERPs在决策者方案选择时的电位变化；(2)采集基于眼动行为数据得到的注意程度；(3)获取决策者方案选择时间得到方案偏好程度。结合这三个数据得到隐性决策行为数据，根据决策者的选择得到显性决策数据，计算不同决策者的决策行为数据之间的关联度得到用户相似度。此方法采集显性决策行为与隐性决策行为，建立决策者-方案之间的相关关系并计算推荐次序，形成基于用户相似度关系的优选方案集。

公开(公告)号：WO2021141921A1

公开(公告)日：2021-07-15

申请号：PCT/US2021/012218

申请日：2021-01-05

Applicant: CLEERLY, INC.

Inventor： MIN, James K. ,  EARLS, James P. ,  RODRIGUES MARQUES, Hugo Miguel

IPC: A61B6/00 ,  A61B6/04 ,  A61B6/10 ,  A61B5/0066 ,  A61B5/0075 ,  A61B5/055 ,  A61B5/7267 ,  A61B5/742 ,  A61B5/7475 ,  A61B6/032 ,  A61B6/037 ,  A61B6/481 ,  A61B6/504 ,  A61B6/5205 ,  A61B8/12 ,  A61B8/14 ,  A61K49/04 ,  G06K2009/363 ,  G06K2209/05 ,  G06K9/3216 ,  G06K9/36 ,  G06K9/6215 ,  G06K9/627 ,  G06K9/6298 ,  G06T2207/10081 ,  G06T2207/10088 ,  G06T2207/10101 ,  G06T2207/10132 ,  G06T2207/20081 ,  G06T2207/30048 ,  G06T2207/30101 ,  G06T7/0012

Abstract: The disclosure herein relates to systems, methods, and devices for medical image analysis, diagnosis, risk stratification, decision making and/or disease tracking. In some embodiments, the systems, devices, and methods described herein are configured to analyze non-invasive medical images of a subject to automatically and/or dynamically identify one or more features, such as plaque and vessels, and/or derive one or more quantified plaque parameters, such as radiodensity, radiodensity composition, volume, radiodensity heterogeneity, geometry, location, and/or the like. In some embodiments, the systems, devices, and methods described herein are further configured to generate one or more assessments of plaque-based diseases from raw medical images using one or more of the identified features and/or quantified parameters.

公开(公告)号：WO2021139448A1

公开(公告)日：2021-07-15

申请号：PCT/CN2020/132596

申请日：2020-11-30

Applicant: 平安科技（深圳）有限公司

Inventor： 徐玲玲 ,  雷晨雨 ,  张国辉 ,  宋晨

IPC: G06K9/62 ,  G06K9/6215 ,  G06K9/6247 ,  G06K9/6256

Abstract: 本申请及人工智能领域,提供了一种基于多个源模型修正新模型的方法、装置以及计算机设备，其中方法包括：将第一训练数据输入至新模型中得到第一当前向量；以及，将第一训练数据分别输入至多个预设的源模型中进行计算，得到对应各源模型的特征向量；并融合计算得到指标向量；计算第一当前向量与指标向量之间的梯度值；根据梯度值校正新模型中的参数。本申请的有益效果：通过将训练数据输入现有的多个源模型中，得到对应的多个特征向量，然后融合计算得到指标向量，然后计算指标向量与新模型得到的当前向量之间的梯度值，通过梯度值校正新模型中的参数。使新模型综合了多个源模型融合后的优点，避免了直接使用融合模型，体积变大，速度变慢的问题。