-
公开(公告)号：US20210279471A1
公开(公告)日：2021-09-09
申请号：US17188194
申请日：2021-03-01
Applicant: NEC Laboratories America, Inc.
Inventor： Yi Yang , Biplob Debnath , Giuseppe Coviello , Oliver Po , Srimat Chakradhar , Yang Gao
Abstract: A computer-implemented method executed by at least one processor for detecting tattoos on a human body is presented. The method includes inputting a plurality of images into a tattoo detector, selecting one or more images of the plurality of images including tattoos, extracting, via a feature extractor, tattoo feature vectors from the tattoos found in the one or more images of the plurality of images including tattoos, applying a deep learning tattoo matching model to determine potential matches between the tattoo feature vectors and preexisting tattoo images stored in a tattoo training database, and generating a similarity score between the tattoo feature vectors and one or more of the preexisting tattoo images stored in the tattoo training database.
-
62.发明申请公开(公告)号：US20200311962A1
公开(公告)日：2020-10-01
申请号：US16814248
申请日：2020-03-10
Applicant: NEC Laboratories America, Inc.
Inventor： Yi Yang , Srimat Chakradhar , Tarang Chugh
Abstract: A computer-implemented method executed by at least one processor for detecting tattoos on a human body is presented. The method includes inputting a plurality of images into a tattoo detection module, selecting one or more images of the plurality of images including tattoos with at least three keypoints, the at least three keypoints having auxiliary information related to the tattoos, manually labeling tattoo locations in the plurality of images including tattoos to create labeled tattoo images, increasing a size of the labeled tattoo images identified to be below a predetermined threshold by padding a width and height of the labeled tattoo images, training two different tattoo detection deep learning models with the labeled tattoo images defining tattoo training data, and executing either the first tattoo detection deep learning model or the second tattoo detection deep learning model based on a performance of a general-purpose graphical processing unit.
-
公开(公告)号：US20200296452A1
公开(公告)日：2020-09-17
申请号：US16815453
申请日：2020-03-11
Applicant: NEC Laboratories America, Inc.
Inventor： Utsav Drolia , Min Feng , Wang-pin Hsiung , Srimat Chakradhar , Oliver Po , Kunal Rao
Abstract: Methods and systems for deploying a video analytics system include determining one or more applications for a security system in an environment, including one or more constraints. Each functional module in a directed graph representation of one or more applications is profiled to generate one or more configurations for each functional module. The nodes of each graph representation represent functional modules of the respective application, and repeated module configurations are skipped. Resource usage for each of the one or more applications is estimated using the one or more configurations of each functional module and the one or more constraints. The one or more applications are deployed in the environment.
-
公开(公告)号：US20200293758A1
公开(公告)日：2020-09-17
申请号：US16810061
申请日：2020-03-05
Applicant: NEC Laboratories America, Inc.
Inventor： Biplob Debnath , Erik Kruus , Murugan Sankaradas , Srimat Chakradhar
Abstract: A computer-implemented method for emulating an object recognizer includes receiving testing image data, and emulating, by employing a first object recognizer, a second object recognizer. Emulating the second object recognizer includes using the first object recognizer to perform object recognition on a testing object from the testing image data to generate data, the data including a feature representation for the testing object, and classifying the testing object based on the feature representation and a machine learning model configured to predict whether the testing object would be recognized by a second object recognizer. The method further includes triggering an action to be performed based on the classification.
-
公开(公告)号：US20200050899A1
公开(公告)日：2020-02-13
申请号：US16522711
申请日：2019-07-26
Applicant: NEC Laboratories America, Inc.
Inventor： Biplob Debnath , Debayan Deb , Srimat Chakradhar
Abstract: A method is provided for classifying objects. The method detects objects in one or more images. The method tags each object with multiple features. Each feature describes a specific object attribute and has a range of values to assist with a determination of an overall quality of the one or more images. The method specifies a set of training examples by classifying the overall quality of at least some of the objects as being of an acceptable quality or an unacceptable quality, based on a user's domain knowledge about an application program that takes the objects as inputs. The method constructs a plurality of first-level classifiers using the set of training examples. The method constructs a second-level classifier from outputs of the first-level automatic classifiers. The second-level classifier is for providing a classification for at least some of the objects of either the acceptable quality or the unacceptable quality.
-
Patent Agency Ranking
公开(公告)号：US09904976B2
公开(公告)日：2018-02-27
申请号：US14945548
申请日：2015-11-19
Applicant: NEC Laboratories America, Inc.
Inventor： Yi Yang , Srimat Chakradhar
CPC classification number: G06T1/20 , G06N3/00 , G06N3/0454 , G06N3/063
Abstract: Systems and methods are disclosed for speeding up a computer having a graphics processing unit (GPU) and a general purpose processor (GP-GPU) by decoupling a convolution process for a first matrix into a row part and a column part; expanding the row part into a second matrix; performing matrix multiplication using the second matrix and a filter matrix; and performing reduction on an output matrix.
-
67.发明申请ACCELERATING STREAM PROCESSING BY DYNAMIC NETWORK AWARE TOPOLOGY RE-OPTIMIZATION 审中-公开动态网络加速流程优化拓扑重新优化公开(公告)号：US20160269247A1
公开(公告)日：2016-09-15
申请号：US15069621
申请日：2016-03-14
Applicant: NEC Laboratories America, Inc.
Inventor： Srimat Chakradhar , Naresh Rapolu
CPC classification number: H04L67/12 , H04L43/0894 , H04L45/02 , H04L45/12
Abstract: Aspects of the present disclosure are directed to techniques that improve performance of streaming systems. Accordingly we disclose efficient techniques for dynamic topology re-optimization, through the use of a feedback-driven control loop that substantially solve a number of these performance-impacting problems affecting such streaming systems. More particularly, we disclose a novel technique for network-aware tuple routing using consistent hashing that improves stream flow throughput in the presence of large, run-time overhead. We also disclose methods for dynamic optimization of overlay topologies for group communication operations. To enable fast topology re-optimization with least system disruption, we present a lightweight, fault-tolerant protocol. All of the disclosed techniques were implemented in a real system and comprehensively validated on three real applications. We have demonstrated significant improvement in performance (20% to 200%), while overcoming various compute and network bottlenecks. We have shown that our performance improvements are robust to dynamic changes, as well as complex congestion patterns. Given the importance of stream processing systems and the ubiquity of dynamic network state in cloud environments, our results represent a significant and practical solution to these problems and deficiencies.
Abstract translation: 本公开的方面针对提高流系统的性能的技术。 因此，通过使用反馈驱动的控制回路，我们公开了动态拓扑重新优化的高效技术，其基本上解决了影响这样的流系统的许多影响性能的问题。 更具体地说，我们公开了一种使用一致的散列来实现网络感知元组路由的新型技术，该方法在存在大的运行时间开销的情况下改善了流量吞吐量。 我们还公开了用于组通信操作的覆盖拓扑的动态优化方法。 为了实现快速拓扑重新优化，最少的系统中断，我们提出了一个轻量级的容错协议。 所有公开的技术都在实际系统中实现，并在三个实际应用中得到全面验证。 在克服各种计算和网络瓶颈的同时，我们已经表现出了显着的提升（20％到200％）。 我们已经表明，我们的性能改进对于动态变化以及复杂的拥塞模式是稳健的。 鉴于流处理系统的重要性和云环境中动态网络状态的普及，我们的结果代表了对这些问题和缺陷的重要而实际的解决方案。
-
68.发明申请公开(公告)号：US20160210723A1
公开(公告)日：2016-07-21
申请号：US14945548
申请日：2015-11-19
Applicant: NEC Laboratories America, Inc.
Inventor： Yi Yang , Srimat Chakradhar
CPC classification number: G06T1/20 , G06N3/00 , G06N3/0454 , G06N3/063
Abstract: Systems and methods are disclosed for speeding up a computer having a graphics processing unit (GPU) and a general purpose processor (GP-GPU) by decoupling a convolution process for a first matrix into a row part and a column part; expanding the row part into a second matrix; performing matrix multiplication using the second matrix and a filter matrix; and performing reduction on an output matrix.
Abstract translation: 公开了用于通过将第一矩阵的卷积处理去耦合到行部分和列部分来加速具有图形处理单元（GPU）和通用处理器（GP-GPU）的计算机的系统和方法; 将行部分扩展为第二矩阵; 使用第二矩阵执行矩阵乘法和滤波器矩阵; 并且对输出矩阵进行减少。
-
公开(公告)号：US20150277877A1
公开(公告)日：2015-10-01
申请号：US14667819
申请日：2015-03-25
Applicant: NEC Laboratories America, Inc.
Inventor： Min Feng , Srimat Chakradhar , Linhai Song
IPC: G06F9/45
Abstract: Systems and methods for source-to-source transformation for compiler optimization for many integrated core (MIC) coprocessors, including identifying data dependencies in candidate loops and data elements used in each iteration for arrays, profiling candidate loops to find a proper number m, wherein data transfer and computation for m iterations take an equal amount of time, and creating an outer loop outside the candidate loop, with each iteration of the outer loop executing m iterations of the candidate loop. Data streaming is performed by determining optimum buffer size for one or more arrays and inserting code before the outer loop to create optimum sized buffers, overlapping data transfer between central processing units (CPUs) and MICs with the computation; reusing buffers to reduce memory employed on the MICs, and reusing threads on MICs to repeatedly launch kernels on the MICs for asynchronous data transfer.
Abstract translation: 用于许多集成核心（MIC）协处理器的编译器优化的源到源转换的系统和方法，包括识别用于阵列的每次迭代中使用的候选循环和数据元素中的数据依赖性，分析候选循环以找到适当数量m，其中 m次迭代的数据传输和计算需要等量的时间，并且在候选循环外部创建外部循环，每个外部循环的迭代执行候选循环的m次迭代。 通过确定一个或多个阵列的最佳缓冲区大小并在外部循环之前插入代码来创建最佳大小的缓冲区，在中央处理单元（CPU）与MIC之间重叠数据传输与计算来执行数据流; 重用缓冲区以减少在MIC上使用的存储器，并且在MIC上重复使用线程来重复地在MIC上启动内核以进行异步数据传输。
-
70.发明申请公开(公告)号：US20150212892A1
公开(公告)日：2015-07-30
申请号：US14572153
申请日：2014-12-16
Applicant: NEC Laboratories America, Inc.
Inventor： Cheng-Hong Li , Giuseppe Coviello , Srimat Chakradhar , Arash Rezaei
IPC: G06F11/14
CPC classification number: G06F9/3863 , G06F9/3009 , G06F9/3877 , G06F9/4856 , G06F11/1438 , G06F11/1446 , G06F11/203 , G06F11/2043 , G06F2201/825 , G06F2201/84 , G06F2209/509
Abstract: Methods are provided. A method includes capturing a snapshot of an offload process being executed by one or more many-core processors. The offload process is in signal communication with a host process being executed by a host processor. At least the offload is in signal communication with a monitoring process. The method further includes terminating the offload process on the one or more many-core processors, by the monitor process responsive to a communication between the monitor process and the offload processing being disrupted. The snapshot includes a respective predetermined minimum set of information required to restore a same state of the process as when the snapshot was taken.
Abstract translation: 提供方法。 一种方法包括捕获由一个或多个多核处理器执行的卸载过程的快照。 卸载过程与由主机处理器执行的主机进程进行信号通信。 至少卸载与监视过程进行信号通信。 该方法还包括通过响应于监视进程和卸载处理中断之间的通信的监视进程来终止一个或多个多核处理器上的卸载处理。 快照包括恢复与拍摄快照时相同的处理状态所需的相应的预定最小信息集合。
-
-
-
-
-
-
-
-
-
Search Results
132
records
(took 0.023 seconds)
