专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2021140501A1 MOTION MATCHING ANALYSIS 审中-公开
公开(公告)号：WO2021140501A1
公开(公告)日：2021-07-15
申请号：PCT/IL2020/051325
申请日：2020-12-23
申请人： AGT INTERNATIONAL GMBH , REINHOLD COHN AND PARTNERS
发明人： KAMYSHANSKA, Hanna , DEBES, Christian , TANKOYEU, Ivan , BADER, Thomas
IPC分类号： G06K9/00 , A63B24/00 , A63F13/44 , G09B19/00 , G06K9/00342 , G06K9/6211 , G06T2207/10021 , G06T2207/30244 , G06T7/248 , G06T7/285 , G06T7/74
摘要： A method which comprises, by a processor and memory circuitry: obtaining a first sequence of images of a user, determining correct performance of a pose of the user in the first sequence of images, including - based on at least one image, for each body part of a plurality of body parts of the user, comparing at least one of spatial orientation and position of the body part, or of at least one body node thereof, with at least one of desired spatial orientation and position, - if the comparison does not meet a matching criterion for at least one body part of the user, or of at least one body node thereof, outputting data representative of a mismatch between the at least one body part of the user, or of the at least one body node thereof, and the at least one of desired spatial orientation and position.

2. 发明申请

WO2017199233A1 ANOMALY DETECTION USING SPIKING NEURAL NETWORKS 审中-公开
标题翻译：用SPIKING神经网络进行异常检测
公开(公告)号：WO2017199233A1
公开(公告)日：2017-11-23
申请号：PCT/IL2017/050378
申请日：2017-03-27
申请人： AGT INTERNATIONAL GMBH , REINHOLD COHN AND PARTNERS
发明人： DEBES, Christian , DEISEROTH, Bjorn
IPC分类号： G06N3/04 , G06N3/08
CPC分类号： G06N3/08 , G06K9/00718 , G06K2009/00738 , G06N3/04 , G06N3/049 , G06N3/088
摘要： A method, system and computer program product, for identifying anomalies in a monitored scene, the method comprising: receiving into a spiking neural network sensor readings from a capture device monitoring a scene; and outputting an indication to a change in the scene, wherein the spiking neural network comprises a multiplicity of layers, each of the multiplicity of layers comprising a neuron per substantially each pixel in a sensor capturing the monitored scene, and wherein one or more of the layers comprises a memory-like unit for comparing states occurring at a time difference.
摘要翻译：一种用于识别所监视的场景中的异常的方法，系统和计算机程序产品，所述方法包括：从捕捉设备接收监视场景的传感器读数的尖峰神经网络; 并且向场景中的变化输出指示，其中尖峰神经网络包括多个层，所述多个层中的每一个包括传感器中的每个实质上每个像素捕获所监视的场景的神经元，并且其中一个或多个层包括类似存储器的单元，用于比较在时间差异下发生的状态。

3. 发明申请

WO2017017682A1 DATA FUSION AND CLASSIFICATION WITH IMBALANCED DATASETS BACKGROUND 审中-公开
标题翻译：数据融合和分类与不平等数据库背景
公开(公告)号：WO2017017682A1
公开(公告)日：2017-02-02
申请号：PCT/IL2016/050824
申请日：2016-07-28
申请人： AGT INTERNATIONAL GMBH , REINHOLD COHN AND PARTNERS
发明人： SUKHANOV, Sergey , MERENTITIS, Andreas , DEBES, Christian
IPC分类号： G06F17/30
CPC分类号： G06N99/005 , G06F17/3002 , G06F17/30598
摘要： Method and system for classification in imbalanced datasets within a supervised classification framework. Bootstrap methodology is modified according to k -Nearest Neighbor sampling weights and adaptive target set size principle, to induce weak classifiers from the bootstrap samples in an iterative procedure that results in a set of weak classifiers. A weighted combination scheme is used to adaptively combine the weak classifiers to a strong classifier that achieves good performance for ah classes (reflected as high values for metrics such as G-mean and F-score) as well as good overall accuracy.
摘要翻译：监督分类框架内不平衡数据集分类的方法和系统。 Bootstrap方法根据k-Nearest Neighbor采样权重和自适应目标集大小原理进行修改，以引导一组弱分类器的迭代过程从引导样本中诱导弱分类器。加权组合方案用于将弱分类器自适应地组合到一个强分类器，其实现了对于类的良好性能（反映为诸如G均值和F分数的指标的高值）以及良好的总体精度。

4. 发明申请

WO2015193285A1 FUSION OF DATA FROM HETEROGENEOUS SOURCES 审中-公开
标题翻译：来自异源的数据融合
公开(公告)号：WO2015193285A1
公开(公告)日：2015-12-23
申请号：PCT/EP2015/063431
申请日：2015-06-16
申请人： AGT INTERNATIONAL GMBH
发明人： HUBER, Marco , DEBES, Christian , HEREMANS, Roel , VAN KASTEREN, Tim
IPC分类号： G06N7/00 , G06K9/00
CPC分类号： G06N7/005 , G06F17/30876 , G06K9/00288 , G06K9/3258 , G06K9/6278 , G06K9/6293 , G06N5/02 , H04L67/12
摘要： A system and method to perform multisensory data fusion in a distributed sensor environment for object identification classification. Embodiments of the invention are sensor- agnostic and capable handling a large number of sensors of different types via a gateway which transmits sensor measurements to a fusion engine according to predefined rules. A relation exploiter allows combining sensor measurements with information on object relationships from a knowledge base. Also included in the knowledge base is a travel model for objects, along with a graph generator to enable forecasting of object locations for further correlation of sensor data in object identification. Multiple task managers allow multiple fusion tasks to be performed in parallel for flexibility and scalability of the system.
摘要翻译：一种在分布式传感器环境中执行多感知数据融合的系统和方法，用于对象识别分类。本发明的实施例是传感器无关的，并且能够通过根据预定规则将传感器测量传输到融合引擎的网关来处理不同类型的大量传感器。关系开发者允许将传感器测量与来自知识库的对象关系的信息相结合。知识库中还包括一个对象的旅行模型，以及一个图形生成器，可以对物体位置进行预测，以便在物体识别中进一步检测传感器数据。多个任务管理器允许并行执行多个融合任务，以实现系统的灵活性和可扩展性。

5. 发明申请

WO2015193288A1 FUSION-BASED OBJECT-RECOGNITION 审中-公开
标题翻译：基于融合的对象识别
公开(公告)号：WO2015193288A1
公开(公告)日：2015-12-23
申请号：PCT/EP2015/063436
申请日：2015-06-16
申请人： AGT INTERNATIONAL GMBH
发明人： HUBER, Marco , MERENTITIS, Andreas , HEREMANS, Roel , DEBES, Christian
IPC分类号： G06K9/62
CPC分类号： G06K9/00624 , G06K9/6288 , G06K9/6292 , G06N7/005 , G06T11/60 , G06T2207/10004 , G06T2207/20221
摘要： An object-recognition method and system employing Bayesian fusion algorithm to reiteratively improve probability of correspondence between captured object images and database object images by fusing probability data associated with each of plurality of object image captures.
摘要翻译：使用贝叶斯融合算法的对象识别方法和系统通过融合与多个对象图像捕获中的每一个相关联的概率数据来重复地提高捕获的对象图像与数据库对象图像之间的对应概率。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式