公开(公告)号：EP3008694A1

公开(公告)日：2016-04-20

申请号：EP14738674.2

申请日：2014-06-11

Applicant: Qualcomm Incorporated

Inventor： VRCELJ, Bojan ,  KNOBLAUCH, Daniel ,  KRISHNAMOORTHI, Raghuraman ,  DIAZ SPINDOLA, Serafin

IPC: G06T7/00

CPC classification number: G06K9/78 ,  G06K9/6228 ,  G06T7/75 ,  G06T2207/10028 ,  G06T2207/20081

Abstract: Systems, methods, and devices are described for capturing compact representations of three-dimensional objects suitable for offline object detection, and storing the compact representations as object representation in a database. One embodiment may include capturing frames of a scene, identifying points of interest from different key frames of the scene, using the points of interest to create associated three-dimensional key points, and storing key points associated with the object as an object representation in an object detection database.

Abstract translation: 系统，方法，和装置中描述了用于捕获适于离线物体检测三维物体的简洁表示，和储存简洁表示为数据库中的对象表示。 一个实施例可以包括捕获场景的帧，识别来自场景的不同关键帧感兴趣的点，使用所关注的点来创建关联的三维关键点，并存储与所述对象相关联的关键点，以在对象表示 检测对象数据库。

公开(公告)号：WO2014201176A1

公开(公告)日：2014-12-18

申请号：PCT/US2014/042005

申请日：2014-06-11

Applicant: QUALCOMM INCORPORATED

Inventor： VRCELJ, Bojan ,  KNOBLAUCH, Daniel ,  KRISHNAMOORTHI, Raghuraman ,  DIAZ SPINDOLA, Serafin

IPC: G06T7/00

CPC classification number: G06K9/78 ,  G06K9/6228 ,  G06T7/75 ,  G06T2207/10028 ,  G06T2207/20081

Abstract: Systems, methods, and devices are described for capturing compact representations of three-dimensional objects suitable for offline object detection, and storing the compact representations as object representation in a database. One embodiment may include capturing frames of a scene, identifying points of interest from different key frames of the scene, using the points of interest to create associated three-dimensional key points, and storing key points associated with the object as an object representation in an object detection database.

Abstract translation: 描述了用于捕获适合于离线对象检测的三维对象的紧凑表示，以及将紧凑表示作为对象表示存储在数据库中的系统，方法和设备。 一个实施例可以包括捕获场景的帧，使用感兴趣点来识别来自场景的不同关键帧的兴趣点，以创建相关联的三维关键点，以及将与对象相关联的关键点存储为对象表示。 对象检测数据库。

公开(公告)号：WO2015127246A1

公开(公告)日：2015-08-27

申请号：PCT/US2015/016875

申请日：2015-02-20

Applicant: QUALCOMM INCORPORATED

Inventor： KNOBLAUCH, Daniel ,  MULLONI, Alessandro ,  WAGNER, Daniel

IPC: G06T11/00 ,  G06T15/04

CPC classification number: G06T15/04 ,  G06T5/003 ,  G06T11/001 ,  G06T17/10 ,  G06T17/205 ,  G06T2200/04 ,  G06T2200/28 ,  G06T2210/61

Abstract: One disclosed example method for view independent color equalized 3D scene texturing includes capturing a plurality of keyframes of an object; accessing a 3D representation of the object comprising a surface mesh model for the object, the surface mesh model comprising a plurality of polygons; for each polygon, assigning one of the plurality of keyframes to the polygon based on one or more image quality characteristics associated with a portion of the keyframe corresponding to the polygon; reducing a number of assigned keyframes by changing associations between assigned keyframes; and for each polygon of the surface mesh model having an assigned keyframe: equalizing a texture color of at least a portion of the polygon based at least in part on one or more image quality characteristics of the plurality of keyframes associated with the polygon; and assigning the equalized texture color to the 3D representation of the object.

Abstract translation: 用于查看独立颜色均衡的3D场景纹理的一个公开的示例性方法包括捕获对象的多个关键帧; 访问包括用于对象的表面网格模型的对象的3D表示，所述表面网格模型包括多个多边形; 对于每个多边形，基于与所述多边形对应的关键帧的一部分相关联的一个或多个图像质量特征，将所述多个关键帧中的一个分配给所述多边形; 通过改变分配的关键帧之间的关联来减少多个分配的关键帧; 并且对于具有分配关键帧的表面网格模型的每个多边形：至少部分地基于与多边形相关联的多个关键帧的一个或多个图像质量特性来均衡多边形的至少一部分的纹理颜色; 并将均衡的纹理颜色分配给对象的3D表示。

公开(公告)号：WO2013188308A1

公开(公告)日：2013-12-19

申请号：PCT/US2013/045012

申请日：2013-06-10

Applicant: QUALCOMM INCORPORATED

Inventor： RAMANANDAN, Arvind ,  BRUNNER, Christopher ,  RAMACHANDRAN, Mahesh ,  TYAGI, Abhishek ,  KNOBLAUCH, Daniel ,  CHARI, Murali Ramaswamy

IPC: G06T7/00 ,  H04N5/232

CPC classification number: G06T7/2053 ,  G01S3/786 ,  G06T7/254 ,  G06T2207/10016

Abstract: A mobile device tracks a relative pose between a camera and a target using Vision aided Inertial Navigation System (VINS), that includes a contribution from inertial sensor measurements and a contribution from vision based measurements. When the mobile device detects movement of the target, the contribution from the inertial sensor measurements to track the relative pose between the camera and the target is reduced or eliminated. Movement of the target may be detected by comparing vision only measurements from captured images and inertia based measurements to determine if a discrepancy exists indicating that the target has moved. Additionally or alternatively, movement of the target may be detected using projections of feature vectors extracted from captured images.

Abstract translation: 移动设备使用视觉辅助惯性导航系统（VINS）跟踪摄像机和目标之间的相对姿态，其包括来自惯性传感器测量的贡献以及基于视觉的测量的贡献。 当移动设备检测到目标的移动时，减少或消除了惯性传感器测量对跟踪相机和目标之间的相对姿势的贡献。 可以通过仅从仅捕获图像的测量值和基于惯量的测量结果来比较目标的移动是否存在指示目标已移动的差异，来检测目标的移动。 附加地或替代地，可以使用从捕获的图像提取的特征向量的投影来检测目标的移动。

公开(公告)号：EP3108449A1

公开(公告)日：2016-12-28

申请号：EP15708640.6

申请日：2015-02-20

Applicant: Qualcomm Incorporated

Inventor： KNOBLAUCH, Daniel ,  MULLONI, Alessandro ,  WAGNER, Daniel

IPC: G06T11/00 ,  G06T15/04

CPC classification number: G06T15/04 ,  G06T5/003 ,  G06T11/001 ,  G06T17/10 ,  G06T17/205 ,  G06T2200/04 ,  G06T2200/28 ,  G06T2210/61

Abstract: One disclosed example method for view independent color equalized 3D scene texturing includes capturing a plurality of keyframes of an object; accessing a 3D representation of the object comprising a surface mesh model for the object, the surface mesh model comprising a plurality of polygons; for each polygon, assigning one of the plurality of keyframes to the polygon based on one or more image quality characteristics associated with a portion of the keyframe corresponding to the polygon; reducing a number of assigned keyframes by changing associations between assigned keyframes; and for each polygon of the surface mesh model having an assigned keyframe: equalizing a texture color of at least a portion of the polygon based at least in part on one or more image quality characteristics of the plurality of keyframes associated with the polygon; and assigning the equalized texture color to the 3D representation of the object.

Abstract translation: 一种用于视图独立色彩均衡的3D场景纹理化的公开示例方法包括：捕捉对象的多个关键帧; 访问包括对象的表面网格模型的对象的3D表示，所述表面网格模型包括多个多边形; 对于每个多边形，基于与对应于所述多边形的所述关键帧的一部分相关联的一个或多个图像质量特性，将所述多个关键帧中的一个关键帧分配给所述多边形; 通过改变分配的关键帧之间的关联来减少分配的关键帧的数量; 并且对于具有分配的关键帧的表面网格模型的每个多边形：至少部分地基于与多边形相关联的多个关键帧的一个或多个图像质量特性来均衡多边形的至少一部分的纹理颜色; 并将均衡纹理颜色分配给对象的3D表示。

公开(公告)号：EP2862146A1

公开(公告)日：2015-04-22

申请号：EP13730774.0

申请日：2013-06-10

Applicant: Qualcomm Incorporated

Inventor： RAMANANDAN, Arvind ,  BRUNNER, Christopher ,  RAMACHANDRAN, Mahesh ,  TYAGI, Abhishek ,  KNOBLAUCH, Daniel ,  CHARI, Murali Ramaswamy

IPC: G06T7/00 ,  H04N5/232

CPC classification number: G06T7/2053 ,  G01S3/786 ,  G06T7/254 ,  G06T2207/10016

Abstract: A mobile device tracks a relative pose between a camera and a target using Vision aided Inertial Navigation System (VINS), that includes a contribution from inertial sensor measurements and a contribution from vision based measurements. When the mobile device detects movement of the target, the contribution from the inertial sensor measurements to track the relative pose between the camera and the target is reduced or eliminated. Movement of the target may be detected by comparing vision only measurements from captured images and inertia based measurements to determine if a discrepancy exists indicating that the target has moved. Additionally or alternatively, movement of the target may be detected using projections of feature vectors extracted from captured images.

Abstract translation: 一种移动设备跟踪摄像机和辅助惯性导航系统（VINS）使用一个视觉目标之间的相对姿态，确实包括来自惯性传感器测量作出贡献，并从基于视觉的测量作出贡献。 当移动设备检测到目标的移动，从所述惯性传感器测量的贡献来跟踪相机和目标之间的相对姿势被减少或消除。 目标的运动可以通过从拍摄的图像和基于惯性测量仅测量如果存在差异指示做的目标已经移动比较的眼光来确定矿进行检测。 另外或替代地，目标的运动可以使用从所捕获的图像中提取的特征向量的预测来检测。