公开(公告)号：US20160171768A1

公开(公告)日：2016-06-16

申请号：US14569549

申请日：2014-12-12

Applicant: QUALCOMM Incorporated

Inventor： Youngmin Park ,  Erick Mendez Mendez ,  Gerhard Reitmayr ,  Daniel Wagner ,  Serafin Diaz Spindola

IPC: G06T19/00 ,  G06T7/00 ,  H04N13/00

CPC classification number: G06T19/006 ,  G06T7/13 ,  G06T7/38 ,  G06T2200/04 ,  G06T2207/10012 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/10028 ,  H04N13/15 ,  H04N13/167 ,  H04N2213/003

Abstract: Disclosed are a system, apparatus, and method for depth and color camera image synchronization. Depth and color camera input images are received or otherwise obtained unsynchronized and without associated creation timestamps. An image of one type is compared with an image of a different type to determine a match for synchronization. Matches may be determined according to edge detection or depth coordinate detection. When a match is determined a synchronized pair is formed for processing within an augmented reality output. Optionally the synchronized pair may be transformed to improve the match between the image pair.

Abstract translation: 公开了用于深度和彩色相机图像同步的系统，装置和方法。 深度和彩色相机输入图像被接收或以其他方式获得不同步并且没有关联的创建时间戳。 将一种类型的图像与不同类型的图像进行比较以确定用于同步的匹配。 可以根据边缘检测或深度坐标检测来确定匹配。 当确定匹配时，形成用于在增强现实输出中进行处理的同步对。 可选地，可以变换同步对以改善图像对之间的匹配。

公开(公告)号：US09117263B2

公开(公告)日：2015-08-25

申请号：US13733741

申请日：2013-01-03

Applicant: QUALCOMM Incorporated

Inventor： Ramin Rezaiifar ,  Serafin Diaz Spindola

IPC: G06T5/00 ,  H04N1/409 ,  G06K9/34 ,  G06K9/40

CPC classification number: G06T5/002 ,  G06K9/346 ,  G06K9/40 ,  H04N1/409

Abstract: The noise in an image having text is removed by convolving a shaped kernel centered on a pixel for each pixel in the image. The shaped kernel has a shape configured to identify pixels that are not part of the text. For example, the shaped kernel may be shaped with zeros in a center of the kernel to identify pixels that are not part of the text. A value for the pixel is set to erase the pixel when the resulting convolution value for the pixel is less than a threshold. The process may be repeated multiple times for differently shaped kernels, including kernels of different sizes and different configurations, such as having values greater than one in at least one of a row, column, and diagonal.

Abstract translation: 具有文本的图像中的噪声通过以图像中的每个像素的像素为中心的成形内核进行卷积来去除。 形状的内核具有被配置为识别不是文本的一部分的像素的形状。 例如，形状的内核可以在内核的中心以零形状来识别不是文本的一部分的像素。 当所得到的像素的卷积值小于阈值时，设置像素的值来擦除像素。 对于不同形状的内核，包括不同大小和不同配置的内核，例如在行，列和对角线中的至少一个中具有大于1的值，该过程可以重复多次。

公开(公告)号：US20140327786A1

公开(公告)日：2014-11-06

申请号：US14335123

申请日：2014-07-18

Applicant: QUALCOMM Incorporated

Inventor： Matthew S. Grob ,  Serafin Diaz Spindola ,  Gerald V. Wright, JR. ,  Virginia Walker Keating

IPC: H04N5/232 ,  G06K9/00

CPC classification number: H04N1/00315 ,  G06K9/00 ,  H04L63/083 ,  H04N1/00127 ,  H04N1/00307 ,  H04N1/00408 ,  H04N1/0044 ,  H04N5/23203 ,  H04N5/23222 ,  H04N7/185 ,  H04N2201/0013 ,  H04N2201/0055 ,  H04N2201/0084 ,  H04W4/80

Abstract: A master device images an object device and uses the image to identify the object device. The master device then automatically interfaces with the identified object device, for example, by pairing with the object device. The master device interfaces with a second object device and initiates an interface between the first object device and the second object device. The master device may receive broadcast data from the object device including information about the visual appearance of the object device and use the broadcast data in the identification of the object device. The master device may retrieve data related to the object device and display the related data, which may be display the data over the displayed image of the object device. The master device may provide an interface to control the object device or be used to pass data to the object device.

Abstract translation: 主设备对对象设备进行成像，并使用图像来识别对象设备。 然后，主设备例如通过与对象设备配对，自动与所识别的对象设备进行接口。 主设备与第二对象设备接口并发起第一对象设备和第二对象设备之间的接口。 主设备可以从对象设备接收包括关于对象设备的视觉外观的信息的广播数据，并且在目标设备的识别中使用广播数据。 主设备可以检索与对象设备相关的数据并显示相关数据，该数据可以在对象设备的显示图像上显示数据。 主设备可以提供用于控制对象设备的接口或用于将数据传递到对象设备。

公开(公告)号：US20140270362A1

公开(公告)日：2014-09-18

申请号：US13844692

申请日：2013-03-15

Applicant: QUALCOMM INCORPORATED

Inventor： Seyed Hesameddin Najafi Shoushtari ,  Yanghai Tsin ,  Murali Ramaswamy Chari ,  Serafin Diaz Spindola

IPC: G06K9/00

CPC classification number: G06K9/6204

Abstract: Embodiments include detection or relocalization of an object in a current image from a reference image, such as using a simple and relatively fast and invariant edge orientation based edge feature extraction, then a weak initial matching combined with a strong contextual filtering framework, and then a pose estimation framework based on edge segments. Embodiments include fast edge-based object detection using instant learning with a sufficiently large coverage area for object re-localization. Embodiments provide a good trade-off between computational efficiency of the extraction and matching processes.

Abstract translation: 实施例包括从参考图像中检测或重新定位当前图像中的对象，例如使用简单且相对快速且不变的基于边缘取向的边缘特征提取，然后将弱初始匹配与强的上下文过滤框架组合，然后 基于边缘段的姿态估计框架。 实施例包括使用具有足够大的覆盖区域的即时学习的快速边缘对象检测用于对象重新定位。 实施例提供了提取和匹配过程的计算效率之间的良好折衷。

公开(公告)号：US20140119598A1

公开(公告)日：2014-05-01

申请号：US13830664

申请日：2013-03-14

Applicant: QUALCOMM Incorporated

Inventor： Mahesh Ramachandran ,  Abhishek Tyagi ,  Serafin Diaz Spindola

IPC: G06T11/60 ,  G06K9/38 ,  G06K9/32

CPC classification number: G06T11/60 ,  G06K9/3241 ,  G06K9/38 ,  G06T7/579

Abstract: Method, apparatus, and computer program product for merging multiple maps for computer vision based tracking are disclosed. In one embodiment, a method of merging multiple maps for computer vision based tracking comprises receiving a plurality of maps of a scene in a venue from at least one mobile device, identifying multiple keyframes of the plurality of maps of the scene, and merging the multiple keyframes to generate a global map of the scene.

Abstract translation: 公开了用于合并用于基于计算机视觉的跟踪的多个地图的方法，装置和计算机程序产品。 在一个实施例中，合并用于基于计算机视觉的跟踪的多个地图的方法包括从至少一个移动设备接收场地中的场景的多个地图，识别场景的多个地图中的多个关键帧，并且合并多个 关键帧生成场景的全局映射。