公开(公告)号：US09418480B2

公开(公告)日：2016-08-16

申请号：US13633800

申请日：2012-10-02

Applicant: Frida Issa ,  Pablo Garcia-Morato

Inventor： Frida Issa ,  Pablo Garcia-Morato

IPC: G09G5/00 ,  G06T19/00 ,  G06T7/00 ,  G02B27/01 ,  G06F3/01

CPC classification number: G06T19/006 ,  G02B27/017 ,  G06F3/012 ,  G06T7/74 ,  G06T7/75 ,  G06T19/00 ,  G06T2207/10016

Abstract: The present system provides a tool to estimate the relative pose of a generic object with respect to a camera view-point by processing 2D images from a monocular camera in real-time. The capability of solving the pose estimation problem relies on the robust detection and matching in consecutive image frames of significant visual features belonging to an object of interest. To accomplish this, the system incorporates 3D modeling of the object of interest. In one embodiment, the shape of interest may be approximated by a parametric surface such as a cylinder, sphere, ellipsoid, or even complex non-parametric models. The system can restrain information retrieved at a 2D image level to estimate parameters about the pose. In operation, the accuracy of the 3D pose estimation of the object is a function of the degree of approximation of the selected model and the ability to select and track relevant features across consecutive image frames.

Abstract translation: 本系统提供了一种用于通过从单目相机实时处理2D图像来估计通用对象相对于相机视点的相对姿态的工具。 解决姿态估计问题的能力依赖于属于感兴趣对象的显着视觉特征的连续图像帧中的鲁棒检测和匹配。 为了实现这一点，该系统结合了感兴趣对象的3D建模。 在一个实施例中，感兴趣的形状可以由诸如圆柱体，球体，椭球体或甚至复杂的非参数模型的参数表面近似。 该系统可以限制在2D图像级别检索的信息来估计关于姿态的参数。 在操作中，对象的3D姿态估计的精度是所选模型的近似度的函数以及跨连续图像帧选择和跟踪相关特征的能力的函数。

公开(公告)号：US20140037203A1

公开(公告)日：2014-02-06

申请号：US13564220

申请日：2012-08-01

Applicant: Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

Inventor： Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

IPC: G06T7/00

CPC classification number: G06T7/0081 ,  G06F17/30247

Abstract: An image recognition system in a cloud environment including uploading a plurality of images to the cloud environment, preprocessing the plurality of images, determining image classifiers for each of the plurality of images, extracting the features of each of the plurality of images, storing the images, features and classifiers, determining the image classifiers and key features of an image to be recognized from a multiplatform image device, selecting from the plurality of images, images which have the same classifiers as the image to be identified, matching a best one of the selected images with the image to be identified and displaying the match on a display of the .multi-platform image device. The system further functions by performing one or more of the above in the multiplatform imaging device and/or cloud environment.

Abstract translation: 一种云环境中的图像识别系统，包括将多个图像上载到云环境，预处理多个图像，为多个图像中的每一个确定图像分类器，提取多个图像中的每个图像的特征，存储图像 特征和分类器，确定要从多平台图像设备识别的图像的图像分类器和关键特征，从多个图像中选择具有与要识别的图像相同的分类器的图像，以匹配最佳的 所选择的图像具有要识别的图像，并且在多平台图像设备的显示器上显示匹配。 该系统通过在多平面成像设备和/或云环境中执行上述中的一个或多个来进一步起作用。

公开(公告)号：US20140306953A1

公开(公告)日：2014-10-16

申请号：US14140288

申请日：2013-12-24

Applicant: Pablo Garcia MORATO ,  Frida ISSA

Inventor： Pablo Garcia MORATO ,  Frida ISSA

IPC: G06T17/00

CPC classification number: G06T17/00 ,  G06T15/04 ,  G06T19/20 ,  G06T2219/2016

Abstract: Rendering systems and methods are provided herein, which generate, from received two-dimensional (2D) object information related to an object and 3D model representations, a textured model of the object. The textured model is placed in training scenes which are used to generate various picture sets of the modeled object in the training scenes. These picture sets are used to train image recognition and object tracking computer systems.

Abstract translation: 本文提供了渲染系统和方法，其从接收到的与对象和3D模型表示相关的二维（2D）对象信息生成对象的纹理模型。 纹理模型放置在训练场景中，用于在训练场景中生成建模对象的各种图片集。 这些图片集用于训练图像识别和对象跟踪计算机系统。

公开(公告)号：US20140092132A1

公开(公告)日：2014-04-03

申请号：US13633800

申请日：2012-10-02

Applicant: Frida Issa ,  Pablo Garcia-Morato

Inventor： Frida Issa ,  Pablo Garcia-Morato

IPC: G06T19/00

CPC classification number: G06T19/006 ,  G02B27/017 ,  G06F3/012 ,  G06T7/74 ,  G06T7/75 ,  G06T19/00 ,  G06T2207/10016

Abstract: The present system provides a tool to estimate the relative pose of a generic object with respect to a camera view-point by processing 2D images from a monocular camera in real-time. The capability of solving the pose estimation problem relies on the robust detection and matching in consecutive image frames of significant visual features belonging to an object of interest. To accomplish this, the system incorporates 3D modeling of the object of interest. In one embodiment, the shape of interest may be approximated by a parametric surface such as a cylinder, sphere, ellipsoid, or even complex non-parametric models. The system can restrain information retrieved at a 2D image level to estimate parameters about the pose. In operation, the accuracy of the 3D pose estimation of the object is a function of the degree of approximation of the selected model and the ability to select and track relevant features across consecutive image frames.

Abstract translation: 本系统提供了一种用于通过从单目相机实时处理2D图像来估计通用对象相对于相机视点的相对姿态的工具。 解决姿态估计问题的能力依赖于属于感兴趣对象的显着视觉特征的连续图像帧中的鲁棒检测和匹配。 为了实现这一点，该系统结合了感兴趣对象的3D建模。 在一个实施例中，感兴趣的形状可以由诸如圆柱体，球体，椭球体或甚至复杂的非参数模型的参数表面近似。 该系统可以限制在2D图像级别检索的信息来估计关于姿态的参数。 在操作中，对象的3D姿态估计的精度是所选模型的近似度的函数以及跨连续图像帧选择和跟踪相关特征的能力的函数。

公开(公告)号：US09361540B2

公开(公告)日：2016-06-07

申请号：US13586681

申请日：2012-08-15

Applicant: Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

Inventor： Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

IPC: G06K9/34 ,  G06K9/46

CPC classification number: G06K9/4642

Abstract: An image processing for recognition objectives system including uploading n image to a processor and associated memory via a network, dividing the images into a plurality of sub-images, preprocessing, uploading an image to be recognized from an image source to the processor and associated memory, pre-processing the image to be recognized and a first the plurality of sub-images, extracting key features from the image to be recognized and a first of the plurality of sub-images, comparing the key features of the image and the first of the sub-images, if there is a match, display the match and if there is no match, advance to the next of the plurality of sub-images and repeat until a match is determined.

Abstract translation: 一种用于识别目标系统的图像处理，包括经由网络将n图像上载到处理器和相关联的存储器，将图像划分为多个子图像，预处理，将要从图像源识别的图像上载到处理器和相关联的存储器 对要识别的图像进行预处理，并且首先执行多个子图像，从要识别的图像中提取关键特征和多个子图像中的第一个，比较图像的关键特征和第一个 子图像，如果存在匹配，则显示匹配，并且如果不匹配，则前进到多个子图像中的下一个并重复直到确定匹配。

公开(公告)号：US09135712B2

公开(公告)日：2015-09-15

申请号：US13564220

申请日：2012-08-01

Applicant: Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

Inventor： Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

IPC: G06K9/00 ,  G06T7/00 ,  G06F17/30

CPC classification number: G06T7/0081 ,  G06F17/30247

Abstract: An image recognition system in a cloud environment including uploading a plurality of images to the cloud environment, preprocessing the plurality of images, determining image classifiers for each of the plurality of images, extracting the features of each of the plurality of images, storing the images, features and classifiers, determining the image classifiers and key features of an image to be recognized from a multiplatform image device, selecting from the plurality of images, images which have the same classifiers as the image to be identified, matching a best one of the selected images with the image to be identified and displaying the match on a display of the .multi-platform image device. The system further functions by performing one or more of the above in the multiplatform imaging device and/or cloud environment.

Abstract translation: 一种云环境中的图像识别系统，包括将多个图像上载到云环境，预处理多个图像，为多个图像中的每一个确定图像分类器，提取多个图像中的每个图像的特征，存储图像 特征和分类器，确定要从多平台图像设备识别的图像的图像分类器和关键特征，从多个图像中选择具有与要识别的图像相同的分类器的图像，以匹配最佳的 所选择的图像具有要识别的图像，并且在多平台图像设备的显示器上显示匹配。 该系统通过在多平面成像设备和/或云环境中执行上述中的一个或多个来进一步起作用。

公开(公告)号：US20140309925A1

公开(公告)日：2014-10-16

申请号：US14140405

申请日：2013-12-24

Applicant: Pablo Garcia MORATO ,  Frida ISSA

Inventor： Pablo Garcia MORATO ,  Frida ISSA

IPC: G01C21/20

CPC classification number: G01C21/206

Abstract: A visual positioning system for indoor locations with associated content is provided herein. The system has a map creator and a viewer. The map creator maps the indoor location by acquiring plans thereof, detects paths through the location and associates with the paths frames relating to objects and views of the paths. The viewer allows a user to orient in the indoor location by locating the user with respect to a path. The viewer enhances GPS/WIFI/3G data by matching user-captured images with the frames, and then interactively displaying the user data from the mapped paths with respect to user queries.

Abstract translation: 本文提供了具有相关内容的室内位置的视觉定位系统。 该系统具有地图创建者和观看者。 地图创建者通过获取其计划来映射室内位置，检测通过位置的路径，并且与与路径的对象和视图相关的路径框架相关联。 观众允许用户通过相对于路径定位用户来定向在室内位置。 观众通过将用户捕获的图像与帧匹配来增强GPS / WIFI / 3G数据，然后相对于用户查询从映射的路径交互地显示用户数据。

公开(公告)号：US20140050401A1

公开(公告)日：2014-02-20

申请号：US13586681

申请日：2012-08-15

Applicant: Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

Inventor： Pablo Garcia Morato Fernandez Baillo ,  Frida Issa

IPC: G06K9/34

CPC classification number: G06K9/4642

Abstract: An image processing for recognition objectives system including uploading n image to a processor and associated memory via a network, dividing the images into a plurality of sub-images, preprocessing, uploading an image to be recognized from an image source to the processor and associated memory, pre-processing the image to be recognized and a first the plurality of sub-images, extracting key features from the image to be recognized and a first of the plurality of sub-images, comparing the key features of the image and the first of the sub-images, if there is a match, display the match and if there is no match, advance to the next of the plurality of sub-images and repeat until a match is determined.

Abstract translation: 一种用于识别目标系统的图像处理，包括经由网络将n图像上载到处理器和相关联的存储器，将图像划分为多个子图像，预处理，将要从图像源识别的图像上载到处理器和相关联的存储器 对要识别的图像进行预处理，并且首先执行多个子图像，从要识别的图像中提取关键特征和多个子图像中的第一个，比较图像的关键特征和第一个 子图像，如果存在匹配，则显示匹配，并且如果不匹配，则前进到多个子图像中的下一个并重复直到确定匹配。