公开(公告)号：US08959108B2

公开(公告)日：2015-02-17

申请号：US13276110

申请日：2011-10-18

Applicant: Jose Pio Pereira ,  Shashank Merchant ,  Prashant Ramanathan ,  Sunil Suresh Kulkarni ,  Mihailo Stojancic

Inventor： Jose Pio Pereira ,  Shashank Merchant ,  Prashant Ramanathan ,  Sunil Suresh Kulkarni ,  Mihailo Stojancic

IPC: G06F17/30

CPC classification number: G06F17/30038 ,  G06F17/3002 ,  G06F17/30029 ,  G06F17/30377 ,  G06F17/3056

Abstract: An efficient large scale search system for video and multi-media content using a distributed database and search, and tiered search servers is described. Selected content is stored at the distributed local database and tier1 search server(s). Content matching frequent queries, and frequent unidentified queries are cached at various levels in the search system. Content is classified using feature descriptors and geographical aspects, at feature level and in time segments. Queries not identified at clients and tier1 search server(s) are queried against tier2 or lower search server(s). Search servers use classification and geographical partitioning to reduce search cost. Methods for content tracking and local content searching are executed on clients. The client performs local search, monitoring and/or tracking of the query content with the reference content and local search with a database of reference fingerprints. This shifts the content search workload from central servers to the distributed monitoring clients.

Abstract translation: 描述了使用分布式数据库和搜索以及分层搜索服务器的视频和多媒体内容的高效大规模搜索系统。 所选内容存储在分布式本地数据库和tier1搜索服务器上。 内容匹配频繁查询和频繁的不明查询在搜索系统的各个级别进行缓存。 内容使用特征描述符和地理方面，在特征级别和时间段中进行分类。 在客户端和tier1搜索服务器上未识别的查询被查询到tier2或更低的搜索服务器。 搜索服务器使用分类和地理分区来降低搜索成本。 在客户端执行内容跟踪和本地内容搜索的方法。 客户端用参考内容和具有参考指纹数据库的本地搜索来执行查询内容的本地搜索，监视和/或跟踪。 这将内容搜索工作负载从中央服务器转移到分布式监控客户端。

公开(公告)号：US20120207387A1

公开(公告)日：2012-08-16

申请号：US13432914

申请日：2012-03-28

Applicant: Jose Pio Pereira ,  Mihailo M. Stojancic ,  Shashank Merchant

Inventor： Jose Pio Pereira ,  Mihailo M. Stojancic ,  Shashank Merchant

IPC: G06K9/68

CPC classification number: G06F17/3079 ,  G06F17/30799 ,  G06F17/30805 ,  G06F17/30858 ,  G06F17/30949 ,  G06K9/00711 ,  G06K9/00718 ,  G06K9/00758 ,  G06K2009/00738 ,  G06T7/10 ,  G06T7/11 ,  G06T7/12 ,  G06T7/162 ,  G06T2207/10016 ,  G06T2207/20021

Abstract: A multi-dimensional database and indexes and operations on the multi-dimensional database are described which include video search applications or other similar sequence or structure searches. Traversal indexes utilize highly discriminative information about images and video sequences or about object shapes. Global and local signatures around keypoints are used for compact and robust retrieval and discriminative information content of images or video sequences of interest. For other objects or structures relevant signature of pattern or structure are used for traversal indexes. Traversal indexes are stored in leaf nodes along with distance measures and occurrence of similar images in the database. During a sequence query, correlation scores are calculated for single frame, for frame sequence, and video clips, or for other objects or structures.

Abstract translation: 描述了多维数据库以及关于多维数据库的索引和操作，其包括视频搜索应用或其他类似的序列或结构搜索。 遍历索引利用关于图像和视频序列或关于对象形状的高度辨别信息。 关键点周围的全局和本地签名用于紧凑和鲁棒的检索和感兴趣的图像或视频序列的辨别信息内容。 对于其他对象或结构，模式或结构的相关签名用于遍历索引。 遍历索引与距离测量一起存储在叶节点中，并在数据库中出现类似的图像。 在序列查询期间，计算单帧，帧序列和视频剪辑或其他对象或结构的相关分数。

公开(公告)号：US08171030B2

公开(公告)日：2012-05-01

申请号：US12141337

申请日：2008-06-18

Applicant: Jose Pio Pereira ,  Mihailo M. Stojancic ,  Shashank Merchant

Inventor： Jose Pio Pereira ,  Mihailo M. Stojancic ,  Shashank Merchant

IPC: G06F7/00 ,  G06F17/30

CPC classification number: G06F17/3079 ,  G06F17/30799 ,  G06F17/30805 ,  G06F17/30858 ,  G06F17/30949 ,  G06K9/00711 ,  G06K9/00718 ,  G06K9/00758 ,  G06K2009/00738 ,  G06T7/10 ,  G06T7/11 ,  G06T7/12 ,  G06T7/162 ,  G06T2207/10016 ,  G06T2207/20021

Abstract: A multi-dimensional database and indexes and operations on the multi-dimensional database are described which include video search applications or other similar sequence or structure searches. Traversal indexes utilize highly discriminative information about images and video sequences or about object shapes. Global and local signatures around keypoints are used for compact and robust retrieval and discriminative information content of images or video sequences of interest. For other objects or structures relevant signature of pattern or structure are used for traversal indexes. Traversal indexes are stored in leaf nodes along with distance measures and occurrence of similar images in the database. During a sequence query, correlation scores are calculated for single frame, for frame sequence, and video clips, or for other objects or structures.

Abstract translation: 描述了多维数据库以及关于多维数据库的索引和操作，其包括视频搜索应用或其他类似的序列或结构搜索。 遍历索引利用关于图像和视频序列或关于对象形状的高度辨别信息。 关键点周围的全局和本地签名用于紧凑和鲁棒的检索和感兴趣的图像或视频序列的辨别信息内容。 对于其他对象或结构，模式或结构的相关签名用于遍历索引。 遍历索引与距离测量一起存储在叶节点中，并在数据库中出现类似的图像。 在序列查询期间，计算单帧，帧序列和视频剪辑或其他对象或结构的相关分数。

公开(公告)号：US20120095958A1

公开(公告)日：2012-04-19

申请号：US13276110

申请日：2011-10-18

Applicant: Jose Pio Pereira ,  Shashank Merchant ,  Prashant Ramanathan ,  Sunil Suresh Kulkarni ,  Mihailo Stojancic

Inventor： Jose Pio Pereira ,  Shashank Merchant ,  Prashant Ramanathan ,  Sunil Suresh Kulkarni ,  Mihailo Stojancic

IPC: G06F17/30 ,  G06F15/16

CPC classification number: G06F17/30038 ,  G06F17/3002 ,  G06F17/30029 ,  G06F17/30377 ,  G06F17/3056

Abstract: An efficient large scale search system for video and multi-media content using a distributed database and search, and tiered search servers is described. Selected content is stored at the distributed local database and tier1 search server(s). Content matching frequent queries, and frequent unidentified queries are cached at various levels in the search system. Content is classified using feature descriptors and geographical aspects, at feature level and in time segments. Queries not identified at clients and tier1 search server(s) are queried against tier2 or lower search server(s). Search servers use classification and geographical partitioning to reduce search cost. Methods for content tracking and local content searching are executed on clients. The client performs local search, monitoring and/or tracking of the query content with the reference content and local search with a database of reference fingerprints. This shifts the content search workload from central servers to the distributed monitoring clients.

Abstract translation: 描述了使用分布式数据库和搜索以及分层搜索服务器的视频和多媒体内容的高效大规模搜索系统。 所选内容存储在分布式本地数据库和tier1搜索服务器上。 内容匹配频繁查询和频繁的不明查询在搜索系统的各个级别进行缓存。 内容使用特征描述符和地理方面，在特征级别和时间段中进行分类。 在客户端和tier1搜索服务器上未识别的查询被查询到tier2或更低的搜索服务器。 搜索服务器使用分类和地理分区来降低搜索成本。 在客户端执行内容跟踪和本地内容搜索的方法。 客户端用参考内容和具有参考指纹数据库的本地搜索来执行查询内容的本地搜索，监视和/或跟踪。 这将内容搜索工作负载从中央服务器转移到分布式监控客户端。

公开(公告)号：US20100318515A1

公开(公告)日：2010-12-16

申请号：US12772566

申请日：2010-05-03

Applicant: Prashant Ramanathan ,  Jose Pio Pereira ,  Shashank Merchant

Inventor： Prashant Ramanathan ,  Jose Pio Pereira ,  Shashank Merchant

IPC: G06F17/30 ,  G09G5/00

CPC classification number: G06F17/30038 ,  G06F17/3002 ,  G06F17/3033 ,  G06F17/30598 ,  G06F17/30784 ,  G06F17/30799 ,  G06F17/30864 ,  G06F17/30949 ,  G06K9/00744 ,  G06K9/4671 ,  Y10S707/913

Abstract: The overall architecture and details of a scalable video fingerprinting and identification system that is robust with respect to many classes of video distortions is described. In this system, a fingerprint for a piece of multimedia content is composed of a number of compact signatures, along with traversal hash signatures and associated metadata. Numerical descriptors are generated for features found in a multimedia clip, signatures are generated from these descriptors, and a reference signature database is constructed from these signatures. Query signatures are also generated for a query multimedia clip. These query signatures are searched against the reference database using a fast similarity search procedure, to produce a candidate list of matching signatures. This candidate list is further analyzed to find the most likely reference matches. Signature correlation is performed between the likely reference matches and the query clip to improve detection accuracy.

Abstract translation: 描述了关于许多类视频失真的鲁棒性的可扩展视频指纹识别系统的总体架构和细节。 在该系统中，用于一张多媒体内容的指纹由许多紧凑签名以及遍历哈希签名和相关联的元数据组成。 为多媒体剪辑中发现的特征生成数字描述符，从这些描述符生成签名，并从这些签名构建参考签名数据库。 还为查询多媒体剪辑生成查询签名。 使用快速相似性搜索过程针对参考数据库搜索这些查询签名，以产生匹配签名的候选列表。 进一步分析该候选列表以找到最可能的参考匹配。 在可能的参考匹配和查询剪辑之间执行签名相关性，以提高检测精度。

公开(公告)号：US20080310731A1

公开(公告)日：2008-12-18

申请号：US12141163

申请日：2008-06-18

Applicant: Mihailo M. Stojancic ,  Jose Pio Pereira ,  Shashank Merchant

Inventor： Mihailo M. Stojancic ,  Jose Pio Pereira ,  Shashank Merchant

IPC: G06K9/46

CPC classification number: G06K9/00711

Abstract: Scaleable video sequence processing with various filtering rules is applied to extract dominant features, and generate unique set of signatures based on video content. Video sequence structuring and subsequent video sequence characterization is performed by tracking statistical changes in the content of a succession of video frames and selecting suitable frames for further treatment by region based intra-frame segmentation and contour tracing and description. Compact representative signatures are generated on the video sequence structural level as well as on the selected video frame level, resulting in an efficient video database formation and search.

Abstract translation: 应用具有各种过滤规则的可扩展视频序列处理来提取主要特征，并且基于视频内容生成唯一的签名集合。 通过跟踪一系列视频帧的内容中的统计变化并选择合适的帧以进行基于区域的帧内分割和轮廓跟踪和描述的进一步处理来执行视频序列构造和后续视频序列表征。 在视频序列结构层面以及所选择的视频帧级别上生成紧凑的代表性签名，从而形成和搜索高效的视频数据。

公开(公告)号：US20050289244A1

公开(公告)日：2005-12-29

申请号：US10876775

申请日：2004-06-28

Applicant: Himansu Sahu ,  Pritam De ,  Shashank Merchant ,  Gopi Krishna

Inventor： Himansu Sahu ,  Pritam De ,  Shashank Merchant ,  Gopi Krishna

IPC: G06F15/16 ,  G06F15/173 ,  H04L29/08

Abstract: The invention relates to a method, system, network node and computer program for processing packet data in a communication network, which comprises at least a first network node. In the method a first packet is received at the first network node. In the first network node is assigned for the first packet a chain comprising at least two logical service entities based on at least one service determination rule. A data unit comprising at least part of the first packet is formed. The data unit is processed in at least one logical service entity in the chain and a second packet is transmitted from the first network node comprising data sent by at least one logical service entity in the chain. The benefits of the invention relate to improved flexibility in introducing new value-added service for packet data and improved performance in the first network node.

Abstract translation: 本发明涉及一种在通信网络中处理分组数据的方法，系统，网络节点和计算机程序，其包括至少第一网络节点。 在该方法中，在第一网络节点处接收第一分组。 在第一网络节点被分配用于第一分组基于至少一个服务确定规则的包括至少两个逻辑服务实体的链。 形成包括第一包的至少一部分的数据单元。 数据单元在链中的至少一个逻辑服务实体中处理，并且从第一网络节点发送包括由链中的至少一个逻辑服务实体发送的数据的第二分组。 本发明的优点涉及在为第一网络节点引入用于分组数据的新增值服务和改进的性能方面的改进的灵活性。

公开(公告)号：US06934280B1

公开(公告)日：2005-08-23

申请号：US09565675

申请日：2000-05-04

Applicant: Chi Fai Ho ,  Shashank Merchant ,  Amar Gupta ,  Prabhas Kejriwal ,  Ravi Bail Bhat ,  Guatam Dewan

Inventor： Chi Fai Ho ,  Shashank Merchant ,  Amar Gupta ,  Prabhas Kejriwal ,  Ravi Bail Bhat ,  Guatam Dewan

IPC: H04J3/16 ,  H04L12/24 ,  H04L12/28 ,  H04L12/46 ,  H04L12/56 ,  H04L12/66 ,  H04L29/06

CPC classification number: H04L12/4633 ,  H04L29/06 ,  H04L41/5003 ,  H04L41/5009 ,  H04L41/5022 ,  H04L41/5032 ,  H04L45/00 ,  H04L45/52 ,  H04L47/10 ,  H04L47/20 ,  H04L47/2425 ,  H04L47/2458 ,  H04L47/26 ,  H04L47/32 ,  H04L69/08 ,  H04L69/18 ,  H04L2212/00

Abstract: One embodiment of the present invention receives a plurality of network services, maps the plurality of network services to a single network service at a network layer higher than layer 2, and transmits the single network service. Another embodiment of the present invention receives a single network service at a network layer higher than layer 2, maps the single network service to a plurality of network services, and transmits the plurality of network services.

Abstract translation: 本发明的一个实施例接收多个网络服务，将多个网络服务映射到高于层2的网络层上的单个网络服务，并发送单个网络服务。 本发明的另一实施例在高于层2的网络层接收单个网络服务，将单个网络服务映射到多个网络服务，并发送多个网络服务。

公开(公告)号：US06904043B1

公开(公告)日：2005-06-07

申请号：US09316072

申请日：1999-05-21

Applicant: Shashank Merchant ,  Ching Yu

Inventor： Shashank Merchant ,  Ching Yu

IPC: H04L12/56 ,  H04L29/06 ,  H04L12/28

CPC classification number: H04L49/25 ,  H04L49/205 ,  H04L49/254 ,  H04L49/3027 ,  H04L49/351 ,  H04L49/352 ,  H04L49/354 ,  H04L49/90 ,  H04L49/901 ,  H04L49/9047 ,  H04L49/9073 ,  H04L49/9089 ,  H04L69/22

Abstract: A network switch configured for switching data packets across multiple ports uses an internal memory to store frame headers for processing by decision making logic. The internal memory stores frame headers in a queue configured to store a number of the frame headers for each of the receive ports. A scheduler is included for facilitating the transfer the data from the queues to the decision making logic according to a predetermined priority. The scheduler is also able allocate the time slots in accordance with data traffic at the corresponding receive ports to maximize data throughput.

Abstract translation: 配置用于跨多个端口切换数据包的网络交换机使用内部存储器来存储用于由决策逻辑进行处理的帧头。 内部存储器存储队列中的帧头，被配置为存储每个接收端口的多个帧头。 包括调度器，以便于根据预定优先级将数据从队列传送到决策逻辑。 调度器还可以根据相应接收端口处的数据流量来分配时隙以最大化数据吞吐量。

公开(公告)号：US06744776B1

公开(公告)日：2004-06-01

申请号：US09690890

申请日：2000-10-18

Applicant: Mohan V. Kalkunte ,  Shashank Merchant ,  Phil Simons

Inventor： Mohan V. Kalkunte ,  Shashank Merchant ,  Phil Simons

IPC: H04L1256

CPC classification number: H04L47/2433 ,  H04L47/50

Abstract: A novel method of servicing multiple data queues having different priorities is provided in a network switch. A dequeuing logic circuit services the data queues in a round-robin fashion. Programmable number of data packets is selected from each data queue in each cycle. The dequeuing logic circuit compares the number of data packets selected from a current data queue in a current cycle with the preprogrammed number of data packets set for the current queue, and selects a data packet from the current data queue only if the number of packets selected from the current data queue in the current cycle is less than the preprogrammed number. Selection of a data packet from the current data queue is bypassed, processing a next data queue, if the number of packets selected from the current data queue in the current cycle is not less than the preprogrammed number.

Abstract translation: 在网络交换机中提供了一种维护具有不同优先级的多个数据队列的新颖方法。 出队逻辑电路以循环方式为数据队列提供服务。 在每个循环中从每个数据队列中选择可编程数量的数据包。 出队逻辑电路将当前周期中当前数据队列中选择的数据包数目与为当前队列设置的预编程数据包数量进行比较，只有当选择了分组数量时才从当前数据队列中选择一个数据包 从当前循环中的当前数据队列小于预编程号。 如果从当前循环中的当前数据队列中选择的分组数不少于预编程数，则绕过来自当前数据队列的数据分组的选择，处理下一个数据队列。