公开(公告)号：US20160300145A1

公开(公告)日：2016-10-13

申请号：US15189386

申请日：2016-06-22

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06N5/04 ,  H04N21/25 ,  H04N21/466 ,  H04N21/258 ,  H04N21/45

CPC classification number: G06N5/04 ,  G06F16/957 ,  G06K9/00536 ,  G06K9/00744 ,  G06Q30/0251 ,  G10L15/32 ,  H04H60/37 ,  H04H60/48 ,  H04H60/56 ,  H04H2201/90 ,  H04N7/17318 ,  H04N21/252 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/4532 ,  H04N21/466 ,  H04N21/8106

Abstract: A method and system for identification of a deviation from a periodic behavior pattern in a sequence of multimedia content segments are provided. The method includes generating at least one signature for each multimedia content segment of the sequence of multimedia content segments; comparing at least two of the generated signatures to detect a periodic behavior pattern, wherein the compared at least two signatures represent consecutive multimedia content segments; and comparing at least one signature of at least a subsequently received multimedia content segment to at least one signature representing the periodic behavior pattern to identify a deviation from the periodic behavior pattern.

Abstract translation: 提供了一种用于识别与多媒体内容段的序列中的周期性行为模式的偏差的方法和系统。 该方法包括为多媒体内容片段序列的每个多媒体内容片段生成至少一个签名; 比较所生成的签名中的至少两个以检测周期性行为模式，其中所述比较的至少两个签名表示连续的多媒体内容段; 以及将至少一个随后接收的多媒体内容片段的至少一个签名与表示周期性行为模式的至少一个签名进行比较，以识别与所述周期性行为模式的偏差。

公开(公告)号：US20160239566A1

公开(公告)日：2016-08-18

申请号：US15140977

申请日：2016-04-28

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y Zeevi

IPC: G06F17/30 ,  G10L25/57

CPC classification number: G06F16/685 ,  G06F16/22 ,  G06F16/284 ,  G06F16/285 ,  G06F16/40 ,  G06F16/41 ,  G06F16/61 ,  G06F16/68 ,  G06F16/683 ,  G06F16/7834 ,  G06F16/7844 ,  G06F16/7847 ,  G06K9/00744 ,  G10L13/06 ,  G10L15/063 ,  G10L15/26 ,  G10L15/32 ,  G10L25/54 ,  G10L25/57

Abstract: A system for generating a large-scale database of heterogeneous speech is provided. The system comprises a processor a plurality of independent computation cores configured to generate signatures of a plurality of speech segments; a large scale database configured to maintain a plurality of transcribed multimedia signals; a memory, the memory containing instructions that, when executed by the processor, configure the system to: randomly select a plurality of speech segments from the plurality of multimedia signals, wherein each speech segment of the plurality of speech segments is of a random length; provide the plurality of speech segments to the plurality of independent computation cores for generation of the signatures; collect the signatures from the plurality of independent computation cores; and populate the large-scale database with the plurality of signatures respective of the plurality of multimedia signals.

Abstract translation: 提供了一种用于生成异构语音的大规模数据库的系统。 该系统包括处理器，多个独立的计算核心被配置为生成多个语音段的签名; 配置为维护多个转录多媒体信号的大规模数据库; 存储器，所述存储器包含指令，所述指令在由所述处理器执行时将所述系统配置为：从所述多个多媒体信号中随机选择多个语音段，其中所述多个语音段中的每个语音段具有随机长度; 将所述多个语音段提供给所述多个独立计算核心以产生所述签名; 从多个独立的计算核心收集签名; 并且利用多个多媒体信号中的多个签名填充该大型数据库。

公开(公告)号：US20150379751A1

公开(公告)日：2015-12-31

申请号：US14836254

申请日：2015-08-26

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06T11/60

CPC classification number: G06T11/60 ,  G06F16/9554 ,  G06T2201/005 ,  H04H20/31 ,  H04H2201/50 ,  H04N7/17318 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/466 ,  H04N21/8106

Abstract: A method and system for embedding a code in a multimedia content item are provided. The method comprises identifying multimedia content elements existing in the multimedia content item; generating a new multimedia content element based on the identified existing multimedia content elements; and adding the at least one new multimedia content element to the multimedia content item.

Abstract translation: 提供了一种用于在多媒体内容项中嵌入代码的方法和系统。 该方法包括：识别多媒体内容项中存在的多媒体内容元素; 基于所识别的现有多媒体内容元素生成新的多媒体内容元素; 以及将所述至少一个新的多媒体内容元素添加到所述多媒体内容项目。

公开(公告)号：US09218606B2

公开(公告)日：2015-12-22

申请号：US13874115

申请日：2013-04-30

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06N5/02 ,  G06F17/30 ,  H04N21/466 ,  H04N21/81 ,  H04N21/258 ,  G06Q30/02 ,  H04N7/173 ,  H04N21/2668

CPC classification number: G06Q30/0245 ,  G06F17/30864 ,  G06N5/02 ,  G06Q30/0201 ,  H04N7/17318 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/466 ,  H04N21/8106

Abstract: A system and method for monitoring a brand sentiment through a plurality of different web sources are provided. The method comprises receiving a request to monitor a brand; searching the plurality of web sources for multimedia content elements related to the brand; generating at least one signature for each multimedia content element determined to be related to the brand, wherein each of the at least one generated signatures represents a concept; and correlating the concepts respective of the generated signatures to determine a context of the multimedia content elements determined to be related to the brand, wherein the context is the brand sentiment.

Abstract translation: 提供了一种用于通过多个不同网络源监视品牌情绪的系统和方法。 该方法包括接收对品牌进行监控的请求; 搜索所述多个网页来获取与所述品牌相关的多媒体内容元素; 为确定与品牌相关的每个多媒体内容元素生成至少一个签名，其中所述至少一个生成的签名中的每一个表示概念; 并且将生成的签名相应的概念相关联以确定被确定为与品牌相关的多媒体内容元素的上下文，其中上下文是品牌情感。

公开(公告)号：US20150332154A1

公开(公告)日：2015-11-19

申请号：US14811219

申请日：2015-07-28

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06N5/04

CPC classification number: G06N5/04 ,  G06F16/40 ,  G06Q30/0201 ,  H04H60/37 ,  H04H60/49 ,  H04H60/71 ,  H04H2201/90 ,  H04N7/17318 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/466 ,  H04N21/8106

Abstract: A method and system for identifying social trends are provided. The method includes collecting multimedia content from a plurality of data sources; gathering environmental variables related to the collected multimedia content; extracting visual elements from the collected multimedia content; generating at least one signature for each extracted visual element; generating at least one cluster of visual elements by clustering at least similar signatures generated for the extracted visual elements; correlating environmental variables related to visual elements in the at least one cluster; determining at least one social trend by associating the correlated environmental variables with the at least one cluster.

Abstract translation: 提供了一种识别社会趋势的方法和系统。 该方法包括从多个数据源收集多媒体内容; 收集与收集的多媒体内容相关的环境变量; 从收集的多媒体内容中提取视觉元素; 为每个提取的视觉元素生成至少一个签名; 通过聚集为所提取的可视元素生成的至少类似的签名来生成至少一个视觉元素的集群; 将与至少一个群集中的视觉元素相关的环境变量相关联; 通过将所述相关环境变量与所述至少一个集群相关联来确定至少一个社会趋势。

公开(公告)号：US09104747B2

公开(公告)日：2015-08-11

申请号：US14334903

申请日：2014-07-18

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06F15/18 ,  G06F17/30 ,  G06N5/02

CPC classification number: G06F17/30598 ,  G06F17/3028 ,  G06F17/30705 ,  G06F17/30994 ,  G06N5/025

Abstract: A method and system for signature-based unsupervised clustering of data elements. The method comprises receiving a plurality of clusters; generating a triangular matrix respective of the clusters; generating a signature for each of the clusters; generating a match score between each of two different clusters; storing the match score in a cell of the triangular matrix corresponding to the two clusters; determining whether any of the match scores is above a predefined threshold value; clustering every two clusters that are determined to have a score above a predetermined threshold; and repeating the generation of a triangular matrix respective of the clusters until a single cluster is reached. The system comprises an interface; a processor; a memory for storing at least one cluster; and a memory coupled to the processor, the memory containing instructions that, when executed by the processor, configure the system to perform the steps of the method.

Abstract translation: 一种用于基于签名的无监督聚类数据元素的方法和系统。 该方法包括接收多个簇; 生成各个簇的三角矩阵; 为每个聚类生成签名; 在两个不同的聚类之间产生匹配分数; 将匹配分数存储在对应于两个聚类的三角矩阵的小区中; 确定所述匹配得分是否高于预定阈值; 将被确定为具有高于预定阈值的分数的每两个聚类聚类; 并且重复生成各个簇的三角矩阵，直到达到单个簇。 该系统包括一个接口; 处理器 用于存储至少一个簇的存储器; 以及耦合到所述处理器的存储器，所述存储器包含指令，所述指令在由所述处理器执行时配置所述系统以执行所述方法的步骤。

公开(公告)号：US20150199355A1

公开(公告)日：2015-07-16

申请号：US14638176

申请日：2015-03-04

Applicant: CORTICA, LTD.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06F17/30

CPC classification number: G09B19/0092 ,  H04N7/17318 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/466 ,  H04N21/8106

Abstract: A method and system for identifying a correct orientation of a multimedia content item are presented. The method includes receiving from a user device the multimedia content item; identifying at least one object shown in the multimedia content item; generating by a signature generator system (SGS) at least one signature for the at least one object shown in the multimedia content item; querying, using the at least one generated signature, a deep-content-classification (DCC) system to find at least one concept that matches the at least one object, wherein the querying of the DCC system is performed using the at least one signature generated for each object shown in the multimedia content item; determining a correct orientation of the at least one matching concept; and comparing an orientation of the at least one object to the determined correct orientation to determine if the at least one object is correctly oriented.

Abstract translation: 提出了一种用于识别多媒体内容项目的正确方向的方法和系统。 该方法包括从用户设备接收多媒体内容项; 识别所述多​​媒体内容项目中所示的至少一个对象; 通过签名生成器系统（SGS）生成所述多媒体内容项目中所示的至少一个对象的至少一个签名; 使用所述至少一个生成的签名查询深度内容分类（DCC）系统来找到与所述至少一个对象匹配的至少一个概念，其中使用所生成的所述至少一个签名来执行所述DCC系统的查询 对于多媒体内容项目中所示的每个对象; 确定所述至少一个匹配概念的正确取向; 以及将所述至少一个对象的方向与所确定的正确方向进行比较，以确定所述至少一个对象是否被正确定向。

公开(公告)号：US20150199336A1

公开(公告)日：2015-07-16

申请号：US14638210

申请日：2015-03-04

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06F17/28

Abstract: A system and method for recognizing characters embedded in multimedia content are provided. The method includes extracting at least one image of at least one character from a received multimedia content item; identifying a natural language character corresponding to the at least one image of the at least one character, wherein the identification is performed by a deep content classification (DCC) system; and storing the identified natural language character in a data warehouse.

Abstract translation: 提供了一种用于识别嵌入在多媒体内容中的字符的系统和方法。 该方法包括从接收到的多媒体内容项中提取至少一个角色的至少一个图像; 识别与所述至少一个字符的所述至少一个图像相对应的自然语言字符，其中所述标识由深度内容分类（DCC）系统执行; 并将所识别的自然语言字符存储在数据仓库中。

公开(公告)号：US20150186405A1

公开(公告)日：2015-07-02

申请号：US14643694

申请日：2015-03-10

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06F17/30

CPC classification number: G06F17/30156 ,  G06F17/3002 ,  G06F17/30023 ,  G06F17/30321 ,  G06F17/30598 ,  Y10S707/99943 ,  Y10S707/99948

Abstract: A system and method for generating a concept database respective of a plurality of multimedia data elements (MMDEs). The method comprises: generating a plurality of items from a received MMDE of the plurality of MMDEs; determining the items that are of interest for signature generation; generating at least one signature responsive to at least one item of interest of the received MMDE; clustering a plurality of signatures received from the signature generator responsive of the plurality of MMDEs; reducing the number of signatures in each cluster to a create a signature reduced cluster (SRC) of the cluster; associating metadata with the SRC to a concept structure comprised of a plurality of SRCs and their associated metadata; and generating at least one index for mapping the received MMDE to at least one concept structure, wherein the concept database includes concept structures and the generated indices for the plurality of MMDEs.

Abstract translation: 一种用于生成多个多媒体数据元素（MMDE）的概念数据库的系统和方法。 该方法包括：从所接收的多个MMDE的MMDE中产生多个项目; 确定签名生成感兴趣的项目; 响应于所接收的MMDE的至少一个兴趣项产生至少一个签名; 根据多个MMDE对从签名生成器接收的多个签名进行聚类; 将每个集群中的签名数量减少到创建集群的签名减少集群（SRC）; 将元数据与SRC相关联到由多个SRC及其相关元数据组成的概念结构; 以及生成用于将所接收的MMDE映射到至少一个概念结构的至少一个索引，其中所述概念数据库包括所述多个MMDE的概念结构和生成的索引。

公开(公告)号：US20150161653A1

公开(公告)日：2015-06-11

申请号：US14621661

申请日：2015-02-13

Applicant: Cortica, Ltd.

Inventor： Igal Raichelgauz ,  Karina Odinaev ,  Yehoshua Y. Zeevi

IPC: G06Q30/02

CPC classification number: G06Q30/0242 ,  G06F16/957 ,  G06Q30/0251 ,  H04N7/17318 ,  H04N21/25891 ,  H04N21/2668 ,  H04N21/466 ,  H04N21/8106

Abstract: A system and method for generating an advertisement effectiveness performance score of a multimedia content element displayed in a webpage are provided. The method includes determining an advertisement context of an advertisement; analyzing metadata associated with at least one prior advertisement to determine a prior advertisement success score; determining a multimedia context of the multimedia content element displayed in the webpage; matching the advertisement context to the multimedia context to determine a context matching score; and generating an advertisement effectiveness score at least based on the prior advertisement success score and the context matching score

Abstract translation: 提供了一种用于产生网页中显示的多媒体内容元素的广告效果性能分数的系统和方法。 该方法包括确定广告的广告语境; 分析与至少一个先前广告相关联的元数据，以确定先前的广告成功分数; 确定网页中显示的多媒体内容元素的多媒体上下文; 将广告上下文匹配到多媒体上下文以确定上下文匹配分数; 以及至少基于先前的广告成功分数和上下文匹配分数来生成广告有效性得分