公开(公告)号：WO2015031350A1

公开(公告)日：2015-03-05

申请号：PCT/US2014/052689

申请日：2014-08-26

Applicant: QUALCOMM INCORPORATED

Inventor： XU, Lei ,  ZHOU, Bo ,  CASTELLOE, Michael W. ,  QUAN, Shuxue ,  ZHANG, Xinping ,  DU, Junchen ,  HARTHATTU, Ashwath ,  GUO, Feng ,  QI, Yingyong

IPC: G06K9/00

CPC classification number: G06F12/0802 ,  G06F2212/60 ,  G06K9/00973

Abstract: A method for memory utilization by an electronic device is described. The method includes transferring a first portion of a first decision tree and a second portion of a second decision tree from a first memory to a cache memory. The first portion and second portion of each decision tree are stored contiguously in the first memory. The first decision tree and second decision tree are each associated with a different feature of an object detection algorithm. The method also includes reducing cache misses by traversing the first portion of the first decision tree and the second portion of the second decision tree in the cache memory based on an order of execution of the object detection algorithm.

Abstract translation: 描述了一种由电子设备进行存储器利用的方法。 该方法包括将第一决策树的第一部分和第二决策树的第二部分从第一存储器传送到高速缓冲存储器。 每个决策树的第一部分和第二部分被连续地存储在第一存储器中。 第一决策树和第二决策树各自与对象检测算法的不同特征相关联。 该方法还包括通过基于对象检测算法的执行顺序遍历高速缓冲存储器中的第一决策树的第一部分和第二决策树的第二部分来减少高速缓存未命中。

公开(公告)号：WO2015006014A1

公开(公告)日：2015-01-15

申请号：PCT/US2014/042185

申请日：2014-06-12

Applicant: QUALCOMM INCORPORATED

Inventor： GUO, Feng ,  BI, Ning ,  QI, Yingyong

IPC: G06T7/00

CPC classification number: G06T15/00 ,  G06K9/00208 ,  G06T7/593 ,  G06T2207/30201

Abstract: A method for three-dimensional face generation is described. An inverse depth map is calculated based on a depth map and an inverted first matrix. The inverted first matrix is generated from two images in which pixels are aligned vertically and differ horizontally. The inverse depth map is normalized to correct for distortions in the depth map caused by image rectification. A three-dimensional face model is generated based on the inverse depth map and one of the two images.

Abstract translation: 描述用于三维面部生成的方法。 基于深度图和倒置的第一矩阵计算反深度图。 反转的第一矩阵从两个图像生成，其中像素垂直对齐并且水平不同。 归一化反相深度图以校正由图像校正导致的深度图中的失真。 基于逆深度图和两个图像之一生成三维面部模型。

公开(公告)号：WO2014197475A1

公开(公告)日：2014-12-11

申请号：PCT/US2014/040701

申请日：2014-06-03

Applicant: QUALCOMM INCORPORATED

Inventor： QI, Yingyong ,  KANDHADAI, Ananthapadmanabhan Arasanipalai ,  MOHAMED, Magdi Abuelgasim ,  QUAN, Shuxue ,  XIE, Binglong

IPC: G06F9/44 ,  G06F9/45

CPC classification number: G06F8/35 ,  G06F8/34 ,  G06F8/42 ,  G06F8/44

Abstract: A method includes capturing an image of a scene that includes a diagram. The method further includes applying functional block recognition rules to image data of the image to recognize functional blocks of the diagram. The functional blocks include at least a first functional block associated with a first computer operation. The method further includes determining whether the functional blocks comply with functional block syntax rules. A functional graph is computer-generated based on the functional blocks complying with the functional block syntax rules. The functional graph corresponds to the diagram, and the functional graph includes the functional blocks.

Abstract translation: 一种方法包括捕获包括图的场景的图像。 该方法还包括将功能块识别规则应用于图像的图像数据以识别该图的功能块。 功能块至少包括与第一计算机操作相关联的第一功能块。 该方法还包括确定功能块是否符合功能块语法规则。 基于符合功能块语法规则的功能块，计算机生成功能图。 功能图对应于图，功能图包括功能块。

公开(公告)号：WO2014116485A1

公开(公告)日：2014-07-31

申请号：PCT/US2014/011724

申请日：2014-01-15

Applicant: QUALCOMM INCORPORATED

Inventor： GAO, Dashan ,  ZHONG, Xin ,  QI, Yingyong ,  BI, Ning

IPC: G06T7/20

CPC classification number: G06T7/20 ,  G06T2207/10016 ,  G06T2207/20021 ,  G06T2207/20104 ,  G06T2207/30241

Abstract: A method for detecting and tracking a target object is described. The method includes performing motion-based tracking for a current video frame by comparing a previous video frame and the current video frame. The method also includes selectively performing object detection in the current video frame based on a tracked parameter.

Abstract translation: 描述用于检测和跟踪目标对象的方法。 该方法包括通过比较先前视频帧和当前视频帧来对当前视频帧执行基于运动的跟踪。 该方法还包括基于跟踪参数在当前视频帧中选择性地执行对象检测。

公开(公告)号：WO2012058481A1

公开(公告)日：2012-05-03

申请号：PCT/US2011/058166

申请日：2011-10-27

Applicant: QUALCOMM INCORPORATED ,  QI, Yingyong

Inventor： QI, Yingyong

IPC: G06K9/46

CPC classification number: G06K9/4671 ,  G06K9/6247

Abstract: Present embodiments contemplate systems, apparatus, and methods to improve feature generation for object recognition. Particularly, present embodiments contemplate excluding and/or modifying portions of images corresponding to dispersed pixel distributions. By excluding and/or modifying these regions within the feature generation process, fewer unfavorable features are generated and computation resources may be more efficiently employed.

Abstract translation: 本发明的实施例考虑用于改善用于对象识别的特征生成的系统，装置和方法。 特别地，本发明实施例考虑排除和/或修改与分散的像素分布相对应的图像部分。 通过在特征生成过程中排除和/或修改这些区域，生成更少的不利特征，并且可以更有效地使用计算资源。

公开(公告)号：WO2008086049A1

公开(公告)日：2008-07-17

申请号：PCT/US2008/050025

申请日：2008-01-02

Applicant: QUALCOMM Incorporated ,  WANG, Haohong ,  LI, Hsiang-Tsun ,  MANJUNATH, Sharath ,  QI, Yingyong

Inventor： WANG, Haohong ,  LI, Hsiang-Tsun ,  MANJUNATH, Sharath ,  QI, Yingyong

IPC: G06T15/10 ,  H04N13/00

CPC classification number: G06T15/10 ,  H04N13/275

Abstract: The rendering of 3D video images on a stereo-enabled display (e.g., stereoscopic or autostereoscopic display) is described. The process includes culling facets facing away from a viewer, defining foreground facets for Left and Right Views and common background facets, determining lighting for these facets, and performing screen mapping and scene rendering for one view (e.g., Right View) using computational results for facets of the other view (i.e., Left View). In one embodiment, visualization of images is provided on the stereo-enabled display of a low-power device, such as mobile phone, a computer, a video game platform, or a Personal Digital Assistant (PDA) device.

Abstract translation: 描述了在支持立体声的显示器上（例如，立体显示或自动立体显示）上的3D视频图像的呈现。 该过程包括面向远离观察者的剔除面，定义用于左视图和右视图以及常见背景面的前景面，确定这些面的照明，以及使用计算结果为一个视图（例如，右视图）执行屏幕映射和场景渲染 另一个视图的方面（即左视图）。 在一个实施例中，在诸如移动电话，计算机，视频游戏平台或个人数字助理（PDA）设备的低功率设备的支持立体声的显示器上提供图像的可视化。

公开(公告)号：WO2008019262A3

公开(公告)日：2008-02-14

申请号：PCT/US2007/074889

申请日：2007-07-31

Applicant: QUALCOMM INCORPORATED ,  QI, Yingyong

Inventor： QI, Yingyong

IPC: H04N7/50

Abstract: Techniques for performing mesh-based video compression/decompression with domain transformation are described. A video encoder partitions an image into meshes of pixels, processes the meshes of pixels to obtain blocks of prediction errors, and codes the blocks of prediction errors to generate coded data for the image. The meshes may have arbitrary polygonal shapes and the blocks may have a predetermined shape, e.g., square. The video encoder may process the meshes of pixels to obtain meshes of prediction errors and may then transform the meshes of prediction errors to the blocks of prediction errors. Alternatively, the video encoder may transform the meshes of pixels to blocks of pixels and may then process the blocks of pixels to obtain the blocks of prediction errors. The video encoder may also perform mesh-based motion estimation to determine reference meshes used to generate the prediction errors.

公开(公告)号：WO2007087619A2

公开(公告)日：2007-08-02

申请号：PCT/US2007/061084

申请日：2007-01-25

Applicant: QUALCOMM INCORPORATED ,  QI, Yingyong

Inventor： QI, Yingyong

IPC: H04N7/26 ,  G06T7/20

CPC classification number: H04N19/527 ,  G06T7/223

Abstract: In a video system a method and/or apparatus to process video blocks comprising: the generation of at least one set of projections for a video block in a first frame, and the generation of at least one set of projections for a video block in a second frame, The at least one set of projections from the first frame are compared to the at least one set of projections from the second frame. The result of the comparison produces at least one projection correlation error (PCE) value.

Abstract translation: 在视频系统中，一种用于处理视频块的方法和/或装置包括：在第一帧中生成针对视频块的至少一组投影，以及生成至少一组 对第二帧中的视频块的投影进行比较。将来自第一帧的至少一组投影与来自第二帧的至少一组投影进行比较。 比较结果至少产生一个投影相关误差（PCE）值。

公开(公告)号：WO2007085004A2

公开(公告)日：2007-07-26

申请号：PCT/US2007/060804

申请日：2007-01-19

Applicant: QUALCOMM INCORPORATED ,  LI, Jingqiang ,  NOYES, Ying Xie ,  QI, Yingyong

Inventor： LI, Jingqiang ,  NOYES, Ying Xie ,  QI, Yingyong

IPC: G06T5/00

CPC classification number: H04N5/23248 ,  G06T7/20 ,  G06T7/30

Abstract: The registration of images comprising generating a plurality of projections from a base frame and generating a plurality of projections from a movement frame. Comparing a set of projections from the base frame, with a second set of projections from the movement frame, and generating a global motion vector estimate to add to the base frame.

Abstract translation: 图像的配准包括从基架产生多个突起并从运动框架产生多个突起。 将来自基本帧的一组投影与来自移动帧的第二组投影进行比较，并且生成全局运动矢量估计以添加到基本帧。

公开(公告)号：WO2004084143A1

公开(公告)日：2004-09-30

申请号：PCT/US2004/008285

申请日：2004-03-17

Applicant: QUALCOMM INCORPORATED ,  QI, Yingyong ,  TIAN, Yushi ,  BI, Ning

Inventor： QI, Yingyong ,  TIAN, Yushi ,  BI, Ning

IPC: G06T15/10

CPC classification number: G06T11/40 ,  G06T15/005 ,  G06T2210/12

Abstract: This disclosure describes an apparatus, such as a wireless communication device, that applies a direct evaluation technique to render triangles for the 3D graphical environment. The apparatus includes a rendering engine that defines a rectangular area of pixels, referred to as a bounding box, that bounds the area to be rendered. The rendering engine evaluates coordinates associated with the pixels of the rectangular area to selectively render those pixels that fall within the triangular area. The direct evaluation triangle rendering algorithm may require fewer complex operations than the more computationally intensive interpolation process employed by other systems. As a result, the apparatus may present a 3D graphical environment while preserving as much as possible the available power.

Abstract translation: 本公开描述了一种诸如无线通信设备的装置，其应用直接评估技术来渲染3D图形环境的三角形。 该装置包括描绘引擎，该引擎定义了限定要渲染的区域的被称为边界框的像素的矩形区域。 渲染引擎评估与矩形区域的像素相关联的坐标，以选择性地渲染落在三角形区域内的像素。 直接评估三角形渲染算法可能需要比其他系统使用的计算密集型插值过程更少的复杂操作。 结果，该装置可以呈现3D图形环境，同时尽可能多地保留可用功率。