公开(公告)号：WO2019045889A1

公开(公告)日：2019-03-07

申请号：PCT/US2018/042222

申请日：2018-07-16

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： TRAN, Quoc-Huy ,  SALEM, Mohammed E. F. ,  ZIA, Muhammad Zeeshan ,  VERNAZA, Paul ,  CHANDRAKER, Manmohan

IPC: G06T7/55 ,  G06T7/11 ,  G06T17/10 ,  G06N3/08

CPC classification number: G06K9/00798 ,  G06K9/00785 ,  G06K9/00825 ,  G06K9/46 ,  G06K9/4628 ,  G06K9/6262 ,  G06K9/6274 ,  G06K9/6276 ,  G06K9/629 ,  G06K9/66 ,  G06N3/0454 ,  G06N3/08 ,  G06T17/05 ,  G06T2200/08 ,  G06T2207/20081

Abstract: Systems and methods for correspondence estimation and flexible ground modeling include communicating two-dimensional (2D) images of an environment to a correspondence estimation module (200), including a first image (102a) and a second image (102b) captured by an image capturing device (100). First features, including geometric features and semantic features, are hierarchically extract from the first image with a first convolutional neural network (CNN) according to activation map weights (210a), and second features, including geometric features and semantic features, are hierarchically extracted from the second image with a second CNN according to the activation map weights (210b). Correspondences between the first features and the second features are estimated, including hierarchical fusing of geometric correspondences and semantic correspondences (220). A 3-dimensional (3D) model of a terrain is estimated using the estimated correspondences belonging to the terrain surface (300). Relative locations of elements and objects in the environment are determined according to the 3D model of the terrain (300). A user is notified of the relative locations (400).

公开(公告)号：WO2019032277A1

公开(公告)日：2019-02-14

申请号：PCT/US2018/043405

申请日：2018-07-24

Applicant: NEC LABORATORIES AMERICA, INC. ,  NEC CORPORATION

Inventor： RHEE, Junghwan ,  WU, Zhenyu ,  KORTS-PARN, Lauri ,  JEE, Kangkook ,  LI, Zhichun ,  SETAYESHFAR, Omid

IPC: G06F21/10 ,  G06F21/56 ,  G06F21/60

Abstract: Systems and methods are disclosed for enhancing cybersecurity in a computer system by detecting safeness levels of executables. An installation lineage of an executable is identified in which entities forming the installation lineage include at least an installer of the monitored executable, and a network address from which the executable is retrieved. Each entity of the entities forming the installation lineage is individually analyzed using at least one safeness analysis. Results of the at least one safeness analysis of each entity are inherited by other entities in the lineage of the executable. A backtrace result for the executable is determined based on the inherited safeness evaluation of the executable. A total safeness of the executable, based on at least the backtrace result, is evaluated against a set of thresholds to detect a safeness level of the executable. The safeness level of the executable is output on a display screen.

公开(公告)号：WO2019027775A1

公开(公告)日：2019-02-07

申请号：PCT/US2018/043803

申请日：2018-07-26

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： SUNDARESAN, Karthikeyan ,  CHAI, Eugene ,  CHAKRABORTY, Ayon ,  RANGARAJAN, Sampath

IPC: H04W84/06 ,  H04W92/10

Abstract: A system for implementing a wireless communication network is provided. The system includes a plurality of unmanned aerial vehicles (UAVs) forming a wireless multi-hop mesh network constituting a backhaul. A given one of the UAVs includes a radio access network (RAN) agent configured to determine at least one UAV configuration for optimized coverage of one or more user equipment (UE) devices in a terrestrial zone, a haul agent configured to coordinate an optimization of the backhaul based at least in part on the at least one UAV configuration determined by the RAN agent, and a core agent configured to implement a distributed core architecture among the plurality of UAVs. The system further includes a controller configured to control the plurality of UAVs based on information received from at least one of the agents.

公开(公告)号：WO2018231671A2

公开(公告)日：2018-12-20

申请号：PCT/US2018/036821

申请日：2018-06-11

Applicant: NEC LABORATORIES AMERICA, INC. ,  NEC CORPORATION

Inventor： YAN, Tan ,  CHEN, Haifeng ,  YASUHIRO, Ajiro

IPC: G06Q20/40 ,  G06Q20/38

CPC classification number: G06Q20/4016 ,  G06Q20/1085 ,  G06Q20/3224 ,  G06Q20/405 ,  G06Q40/02 ,  H04L67/22 ,  H04W4/029

Abstract: A system, method, and computer program product are provided for suspicious remittance detection for a set of users. The method includes detecting (410), by a processor, unrealistic user location movements, based on login activities and remittance activities. The method includes detecting (420), by the processor, abnormal user remittance behavior based on account activities and the remittance activities by detecting any users who are silent for a threshold period of time and thereafter remit an amount of money greater than a threshold money amount. The method includes detecting (430), by the processor, abnormal overall user behavior, based a joint user profile determined across all users from the login activities, the remittance activities, and the account activities. The method includes aggregating (440), by the processor, detection results to generate a final list of suspicious transactions. The method includes performing (450), by the processor, loss preventative actions for each of the suspicious transactions in the final list.

公开(公告)号：WO2018182829A1

公开(公告)日：2018-10-04

申请号：PCT/US2018/014630

申请日：2018-01-22

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： ZONG, Bo ,  ZHANG, Hui

IPC: G05B23/02

CPC classification number: G06N20/00 ,  C07F1/00 ,  C07F1/08 ,  C07F3/003 ,  C07F3/02 ,  C07F3/06 ,  C07F5/003 ,  C07F7/003 ,  C07F7/28 ,  C07F9/005 ,  C07F11/005 ,  C07F13/005 ,  C07F15/0046 ,  C07F15/025 ,  C07F15/045 ,  C07F15/065 ,  C23C16/18 ,  C23C16/45553 ,  G05B23/0218 ,  G06F16/951 ,  G06N5/045 ,  H01L21/02205

Abstract: Systems and methods for automatically generating a set of meta-parameters used to train invariant-based anomaly detectors are provided. Data is transformed into a first set of time series data and a second set of time series data. A fitness threshold search is performed on the first set of time series data to automatically generate a fitness threshold, and a time resolution search is performed on the set of second time series data to automatically generate a time resolution. A set of meta-parameters including the fitness threshold and the time resolution are sent to one or more user devices across a network to govern the training of an invariant-based anomaly detector.

公开(公告)号：WO2018085749A1

公开(公告)日：2018-05-11

申请号：PCT/US2017/060103

申请日：2017-11-06

Applicant: NEC LABORATORIES AMERICA, INC

Inventor： VERNAZA, Paul ,  CHANDRAKER, Manmohan

IPC: G06N3/08 ,  G06N3/04

Abstract: Systems and methods for training semantic segmentation. Embodiments of the present invention include predicting semantic labeling of each pixel in each of at least one training image using a semantic segmentation model. Further included is predicting semantic boundaries at boundary pixels of objects in the at least one training image using a semantic boundary model concurrently with predicting the semantic labeling. Also included is propagating sparse labels to every pixel in the at least one training image using the predicted semantic boundaries. Additionally, the embodiments include optimizing a loss function according the predicted semantic labeling and the propagated sparse labels to concurrently train the semantic segmentation model and the semantic boundary model to accurately and efficiently generate a learned semantic segmentation model from sparsely annotated training images.

Abstract translation:

训练语义分割的系统和方法。 本发明的实施例包括使用语义分割模型来预测至少一个训练图像中的每一个中的每个像素的语义标记。 进一步包括在预测语义标签的同时使用语义边界模型预测至少一个训练图像中的对象的边界像素处的语义边界。 还包括使用预测的语义边界将稀疏标签传播到至少一个训练图像中的每个像素。 另外，实施例包括根据预测的语义标签和传播的稀疏标签优化丢失函数，以同时训练语义分段模型和语义边界模型，以从稀疏注释的训练图像准确并高效地生成学习的语义分段模型。

公开(公告)号：WO2018034740A1

公开(公告)日：2018-02-22

申请号：PCT/US2017/040680

申请日：2017-07-05

Applicant: NEC LABORATORIES AMERICA, INC. ,  NEC HONG KONG LIMITED

Inventor： CHANDRAKER, Manmohan ,  CHOI, Wongun ,  LAU, Eric ,  WONG, Elsa ,  CHEN, Guobin

IPC: G06K9/20 ,  G06N3/08

Abstract: A baby detection system and a mass transit surveillance system are provided. The baby detection system includes a camera (110) configured to capture an input image of a subject purported to be a baby and presented at an electronic-gate system. The baby detection system further includes a memory (122) storing a deep learning model configured to perform a baby detection task for an electronic-gate application corresponding to the electronic-gate system. The baby detection system also includes a processor (121) configured to apply the deep learning model to the input image to provide a baby detection result of either a presence or an absence of an actual baby in relation to the subject purported to be the baby. The baby detection task is configured to evaluate one or more different distractor modalities corresponding to one or more different physical spoofing materials to prevent baby spoofing for the baby detection task.

Abstract translation: 提供婴儿检测系统和公共交通监视系统。 婴儿检测系统包括摄像机（110），摄像机（110）被配置为捕获声称为婴儿的对象的输入图像并呈现在电子门系统处。 婴儿检测系统还包括存储器（122），存储器（122）存储深度学习模型，该模型被配置为对与电子门系统相对应的电子门应用执行婴儿检测任务。 婴儿检测系统还包括配置成将深度学习模型应用于输入图像以提供婴儿检测结果的处理器（121），所述婴儿检测结果指示实际婴儿相对于看来是婴儿的对象存在或不存在。 婴儿检测任务被配置为评估与一种或多种不同的物理欺骗材料相对应的一个或多个不同的牵引器形态，以防止婴儿欺骗以用于婴儿检测任务。

公开(公告)号：WO2017177012A1

公开(公告)日：2017-10-12

申请号：PCT/US2017/026370

申请日：2017-04-06

Applicant: NEC LABORATORIES AMERICA, INC

Inventor： XU, Jianwu ,  ZHANG, Ke ,  ZHANG, Hui ,  MIN, Renqiang ,  JIANG, Guofei

IPC: G06F11/34 ,  G06F11/00 ,  G06N3/02

Abstract: Methods for system failure prediction include clustering log files according to structural log patterns. Feature representations of the log files are determined based on the log clusters. A likelihood of a system failure is determined based on the feature representations using a neural network. An automatic system control action is performed if the likelihood of system failure exceeds a threshold.

Abstract translation:

系统故障预测的方法包括根据结构日志模式对日志文件进行聚类。 日志文件的功能表示是根据日志群集确定的。 基于使用神经网络的特征表示来确定系统故障的可能性。 如果系统故障的可能性超过阈值，则执行自动系统控制动作。

公开(公告)号：WO2017177008A1

公开(公告)日：2017-10-12

申请号：PCT/US2017/026365

申请日：2017-04-06

Applicant: NEC LABORATORIES AMERICA, INC

Inventor： COSATTO, Eric ,  MELVIN, Iain ,  GRAF, Hans Peter

IPC: B60W30/095 ,  B60W50/14 ,  B60W40/04

CPC classification number: G06K9/00805 ,  B60Q9/008 ,  G06K9/00201 ,  G06K9/46 ,  G06K9/4628 ,  G06K9/627 ,  G06N3/0454 ,  G06N3/08

Abstract: A video device for predicting driving situations while a person drives a car is presented. The video device includes multi-modal sensors and knowledge data for extracting feature maps, a deep neural network trained with training data to recognize real-time traffic scenes (TSs) from a viewpoint of the car, and a user interface (UI) for displaying the real-time TSs. The real-time TSs are compared to predetermined TSs to predict the driving situations. The video device can be a video camera. The video camera can be mounted to a windshield of the car. Alternatively, the video camera can be incorporated into the dashboard or console area of the car. The video camera can calculate speed, velocity, type, and/or position information related to other cars within the real-time TS. The video camera can also include warning indicators, such as light emitting diodes (LEDs) that emit different colors for the different driving situations.

Abstract translation: 提出了一种用于预测人在驾驶汽车时的驾驶状况的视频装置。 视频设备包括用于提取特征地图的多模态传感器和知识数据，利用训练数据训练的深度神经网络，以从汽车的角度识别实时交通场景（TS），以及用于显示的用户界面（UI） 实时TS。 实时TS与预定的TS进行比较以预测驾驶情况。 视频设备可以是摄像机。 摄像机可以安装在汽车的挡风玻璃上。 或者，摄像机可以集成到汽车的仪表板或控制台区域。 摄像机可以在实时TS内计算与其他车辆有关的速度，速度，类型和/或位置信息。 摄像机还可以包括警告指示器，例如发光二极管（LED），它们针对不同的驾驶情况发出不同的颜色。

公开(公告)号：WO2017177005A1

公开(公告)日：2017-10-12

申请号：PCT/US2017/026362

申请日：2017-04-06

Applicant: NEC LABORATORIES AMERICA, INC

Inventor： COSATTO, Eric ,  MELVIN, Iain ,  GRAF, Hans Peter

IPC: G08G1/01 ,  G08G1/0965 ,  G08G1/0969 ,  G06N99/00

CPC classification number: G06K9/00805 ,  B60Q9/008 ,  G06K9/00201 ,  G06K9/46 ,  G06K9/4628 ,  G06K9/627 ,  G06N3/0454 ,  G06N3/08

Abstract: A computer-implemented method for training a deep neural network to recognize traffic scenes (TSs) from multi-modal sensors and knowledge data is presented. The computer-implemented method includes receiving data from the multi-modal sensors and the knowledge data and extracting feature maps from the multi-modal sensors and the knowledge data by using a traffic participant (TS) extractor to generate a first set of data, using a static objects extractor to generate a second set of data, and using an additional information extractor. The computer-implemented method further includes training the deep neural network, with training data, to recognize the TSs from a viewpoint of a vehicle.

Abstract translation: 提出了一种用于训练深度神经网络以识别来自多模式传感器和知识数据的交通场景（TS）的计算机实现的方法。 该计算机实现的方法包括通过使用流量参与者（TS）提取器从多模式传感器和知识数据接收数据并且从多模式传感器和知识数据中提取特征映射以生成第一组数据 一个静态对象提取器来生成第二组数据，并使用一个附加信息提取器。 计算机实现的方法还包括训练具有训练数据的深度神经网络以从车辆的角度识别TS。