专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2018222253A1 SYSTEM AND METHOD OF DYNAMICALLY FILTERING DEPTH ESTIMATES TO GENERATE A VOLUMETRIC MAP OF A THREE-DIMENSIONAL ENVIRONMENT HAVING AN ADJUSTABLE MAXIMUM DEPTH 审中-公开
公开(公告)号：WO2018222253A1
公开(公告)日：2018-12-06
申请号：PCT/US2018/023958
申请日：2018-03-23
申请人： QUALCOMM INCORPORATED
发明人： SWEET III, Charles Wheeler , WIERZYNSKI, Casimir
IPC分类号： G05D1/00
CPC分类号： G06T7/593 , B64C39/024 , B64C2201/108 , B64C2201/141 , G01C11/00 , G01C15/002 , G01C21/00 , G05D1/0274 , G05D1/102 , G06K9/00 , G06T7/73 , G06T15/08 , G06T17/05 , G06T2207/10028 , G06T2207/10032 , G06T2207/30241 , G06T2215/16 , H04N13/271
摘要： Various systems and methods of dynamically filtering depth estimates to generate a volumetric map of a three-dimensional (3-D) environment having an adjustable maximum depth include obtaining sensor output including depth estimates and pose estimates in a robotic vehicle, detecting a condition that corresponds to an error level in the pose estimates, filtering the depth estimates obtained from the sensor output based on the detected condition, and generating the volumetric map of the 3-D environment using the filtered depth estimates. Filtering depth estimates obtained from the sensor output based on the detected condition may include adjusting a maximum depth parameter for generating the volumetric map. Further embodiments include a robotic vehicle and/or a computing device within a robotic vehicle including a processor configured with processor-executable instructions for controlling the maximum depth of a volumetric map of a 3-D environment.

2. 发明申请

WO2018209969A1 深度图创建方法与系统及图像虚化方法与系统审中-公开
公开(公告)号：WO2018209969A1
公开(公告)日：2018-11-22
申请号：PCT/CN2017/120331
申请日：2017-12-29
申请人：成都通甲优博科技有限责任公司
发明人：龙学军 , 刘勇 , 周剑
IPC分类号： G06T7/593
CPC分类号： G06T3/0012 , G06T7/593 , G06T2207/10012 , H04N5/23232
摘要：本申请公开了深度图创建方法与系统及图像虚化方法与系统，该深度图创建方法包括：获取双目拍摄系统对目标场景进行拍摄后得到的第一图像和第二图像；分别图像对应的特征点，得到第一组特征点和第二组特征点；分别利用第一组特征点和第二组特征点，相应地确定图像对应的支撑点，得到第一组支撑点和第二组支撑点；基于第一组支撑点和第二组支撑点，对第一图像和第二图像进行稠密立体匹配，以确定第一图像和第二图像之间的视差；利用视差确定出与目标场景对应的深度图。本申请能够大幅提升图像之间的立体匹配精度，从而得到图像之间更加准确的视差，基于该视差得到的深度图将具备较高的质量，从而有利于提升后续的图像虚化效果。

3. 发明申请

WO2018187724A1 SYSTEM, METHOD AND SOFTWARE FOR CONVERTING IMAGES CAPTURED BY A LIGHT FIELD CAMERA INTO THREE-DIMENSIONAL IMAGES THAT APPEAR TO EXTEND VERTICALLY ABOVE OR IN FRONT OF A DISPLAY MEDIUM 审中-公开
公开(公告)号：WO2018187724A1
公开(公告)日：2018-10-11
申请号：PCT/US2018/026516
申请日：2018-04-06
申请人： MAXX MEDIA GROUP, LLC
发明人： FREEMAN, Richard, S. , HOLLINGER, Scott, A.
IPC分类号： G06T7/557 , G06T7/593 , H04N13/204
CPC分类号： G06T7/557 , G06T7/593 , G06T2207/10052 , H04N13/232
摘要： System, method and software for producing imagery with enhanced 3D effects that is viewed on a display medium (14). To create the imagery with enhanced 3D effects, a light field camera (17) images a scene that contains at least one subject (20). At a given moment in time, the subject (20) is positioned in an initial orientation. Upon imaging, the light field camera (17) produces light field image data (16) that represents the scene. The light field image data (16) is edited to add enhanced 3D effects. Once edited, the enhanced image data (43) is played on a display medium (12), where the added 3D effects cause the subject to appear, at least in part, to extend vertically above, or in front of, the display medium (14).

4. 发明申请

WO2018134677A1 MULTI-CAMERA SYSTEM AND METHOD OF USE 审中-公开
公开(公告)号：WO2018134677A1
公开(公告)日：2018-07-26
申请号：PCT/IB2018/000076
申请日：2018-01-19
申请人： HANGZHOU ZERO TECHNOLOGY CO., LTD
发明人： QU, Zheng , JIN, Pengxiang , ZHANG, Tong
IPC分类号： G05D1/02
CPC分类号： G05D1/102 , B64C39/024 , B64C2201/123 , B64C2201/127 , B64C2201/141 , B64C2201/146 , B64D47/08 , G06T7/593 , G06T7/74 , G06T2207/10012 , G06T2207/10024 , G06T2207/10032 , G06T2207/30241 , G06T2207/30261 , H04N5/2252 , H04N5/2256 , H04N5/2258 , H04N5/23203 , H04N5/23229 , H04N5/2354 , H04N5/247 , H04N5/332 , H04N7/181 , H04N7/185 , H04N13/243
摘要： A method for operating a system including a plurality of cameras, the method including: selecting a subset of the cameras, determining a subset of pixels captured by the camera subset, determining a pixel depth associated with each pixel of the pixel subset, and controlling system operation based on the pixel depth.

5. 发明申请

WO2018125914A1 METHOD AND DEVICE FOR VISUALLY IMPAIRED ASSISTANCE 审中-公开
公开(公告)号：WO2018125914A1
公开(公告)日：2018-07-05
申请号：PCT/US2017/068532
申请日：2017-12-27
申请人： VASUYANTRA CORP., A DELAWARE CORPORATION
发明人： MAHERIYA, Kiran K. , SRIVASTAVA, Shiva V.
IPC分类号： G06K9/00 , G06T7/00 , G06T11/00 , G06T17/00 , G06T19/00
CPC分类号： G06K9/00671 , G06K9/00214 , G06K9/4628 , G06K9/6267 , G06K9/6273 , G06T7/536 , G06T7/593 , G06T2207/10021 , G06T2207/20084 , G09B21/003 , G09B21/006 , G09B21/007 , G09B21/008
摘要： Device, system, and method of assistance for visually impaired users. The system comprises a plurality of video cameras, often head mounted, computer processors and associated support devices and algorithms configured for computer vision, and a user worn haptic band comprising a plurality (two or more) of distantly spaced haptic transducers. This haptic band is worn such that user's hands are free for other tasks. The system uses its video camera, depth processing algorithms, and object recognition algorithms (hardware and/or software based) to identify a limited number of navigationally important objects. The spatial locations of each object deemed important is output to the user by varying output to the haptic transducers accordingly. The system is configured to identify and report objects as generic objects, identified objects, and potential obstacle objects. The system can also optionally provide audio information or tactile graphics display information pertaining to these objects as well.

6. 发明申请

WO2018118564A1 USING MULTIPLE CAMERAS TO STITCH A CONSOLIDATED 3D DEPTH MAP 审中-公开
公开(公告)号：WO2018118564A1
公开(公告)日：2018-06-28
申请号：PCT/US2017/066034
申请日：2017-12-13
申请人： SONY CORPORATION , SONY INTERACTIVE ENTERTAINMENT INC. , SHINTANI, Peter
发明人： SHINTANI, Peter , GONZALEZ HERNANDEZ, Jose Omar , CLAY, William , ESPINOSA, Pablo Antonio , ANSFIELD, Fred , MOHAPATRA, Bibhudendu , RESCH, Keith , USAMI, Morio
IPC分类号： H04N13/128 , H04N13/271
CPC分类号： H04N13/156 , G01B11/00 , G01B11/245 , G01B2210/58 , G06T7/593 , G06T2207/10012 , H04N5/33 , H04N13/204 , H04N13/271
摘要： Each of plural devices includes a laser emitter and a camera for detecting reflections of laser light emitted by the device, so that plural of the devices can generate their own depth maps showing images within the field of view of their cameras. The resulting 3D depth maps from multiple cameras at potentially arbitrary locations can be aggregated to create a more accurate 3D depth map of the area covered by all the individual cameras. Each device/camera may be assigned a unique identification such as a unique number and a mechanism to identify, either electronically or visually, devices in its field of view. Using this information, the relative locations of the cameras can be calculated for purposes of aggregating the multiple depth maps.

7. 发明申请

WO2018006296A1 SYSTEMS AND METHODS FOR STEREOSCOPIC IMAGING 审中-公开
标题翻译：用于立体成像的系统和方法
公开(公告)号：WO2018006296A1
公开(公告)日：2018-01-11
申请号：PCT/CN2016/088829
申请日：2016-07-06
申请人： SZ DJI TECHNOLOGY CO., LTD.
发明人： YANG, Kang , GAO, Mingming
IPC分类号： H04N13/00
CPC分类号： G06T7/593 , G06K9/6202 , G06K2009/6213 , G06T2207/10012 , G06T2207/20021 , H04N13/128 , H04N13/239 , H04N2013/0081
摘要： Systems for stereoscopic imaging and methods for making and using same. The system can include one or more processors configured to receive a first image and a second image. A block configuration can be selected from a plurality of block configurations having varying shapes and/or sizes and used for block matching. Points on the first image can be matched with corresponding points on the second image using the selected block configurations, thereby identifying a binocular disparity between the images for depth determination. The present systems and methods are suitable for use aboard mobile platforms such as unmanned aerial vehicles (UAV).
摘要翻译：立体成像系统及其制造和使用方法。该系统可以包括被配置为接收第一图像和第二图像的一个或多个处理器。可以从具有不同形状和/或尺寸的多个块配置中选择块配置并用于块匹配。可以使用选择的块配置将第一图像上的点与第二图像上的对应点相匹配，由此识别图像之间的双目视差以进行深度确定。目前的系统和方法适用于诸如无人机（UAV）之类的移动平台上。

8. 发明申请

WO2018004981A1 FEATURE CHARACTERIZATION FROM INFRARED RADIATION 审中-公开
标题翻译：红外辐射特征表征
公开(公告)号：WO2018004981A1
公开(公告)日：2018-01-04
申请号：PCT/US2017/035470
申请日：2017-06-01
申请人： INTEL CORPORATION
发明人： SHAHAR, Amit , LEE, Chia-Hsun , DAVIES, Robert L. , ORSTEN, Sterling
IPC分类号： G06K9/00 , G06K9/32
CPC分类号： G06F3/0421 , G06F3/005 , G06F3/017 , G06F3/0304 , G06F2203/04101 , G06F2203/04105 , G06K9/00335 , G06K9/00375 , G06T5/003 , G06T5/30 , G06T7/593 , G06T2207/10016 , G06T2207/10021 , G06T2207/10048
摘要： Systems, apparatuses, and/or methods to characterize a user feature. For example, and apparatus may include a pattern receiver to receive a feature infrared (IR) pattern corresponding to non-uniform IR radiation reflected by skin of the user feature and an object IR pattern corresponding to IR radiation reflected by an object. The apparatus may further include a filter to generate a modified IR pattern from the object IR pattern and to remove at least a part of the modified IR pattern from feature IR pattern. In addition, the apparatus may include a feature characterizer to characterize the user feature based on the feature IR pattern. In one example, a computing platform may be controlled based on the characterization of the user feature.
摘要翻译：表征用户特征的系统，装置和/或方法。例如，装置可以包括图案接收器，以接收与由用户特征的皮肤反射的非均匀IR辐射对应的特征红外（IR）图案和与由对象反射的IR辐射对应的对象IR图案。该设备可以进一步包括过滤器，以从物体IR图案生成修改的IR图案并从特征IR图案移除修改的IR图案的至少一部分。另外，该装置可以包括特征表征器，以基于特征IR图案来表征用户特征。在一个示例中，可以基于用户特征的表征来控制计算平台。

9. 发明申请

WO2017212928A1 画像処理装置、画像処理方法、および車両审中-公开
标题翻译：图像处理装置，图像处理方法和车辆
公开(公告)号：WO2017212928A1
公开(公告)日：2017-12-14
申请号：PCT/JP2017/019491
申请日：2017-05-25
申请人：ソニー株式会社
发明人：綱島　宣浩
IPC分类号： G06T7/593 , B60R21/00 , G01B11/00 , G01C3/06 , G03B35/18 , H04N7/18
CPC分类号： B60R21/00 , G01B11/00 , G01C3/06 , G03B35/18 , G06T7/593 , H04N7/18
摘要：本技術は、広角カメラにより撮像された広角画像からなるステレオ画像対を用いて、広範囲に存在する物体までの距離を計測することができるようにする画像処理装置、画像処理方法、および車両に関する。本技術の画像処理装置は、第１の画像の撮像時の視野角を分割した各投影角に従い、前記第１の画像を仮想球面上の複数の平面に投影することによって前記複数の基準平面画像を生成する第１の生成部と、撮像範囲が前記第１の画像と重複した第２の画像の撮像時の視野角を分割した各投影角に従い、前記第２の画像を仮想球面上の複数の平面に投影することによって前記複数の参照平面画像を生成する第２の生成部と、生成された前記複数の基準平面画像と参照平面画像との対応する画像対を用いたステレオ画像処理を行うことにより、前記基準画像上の物体までの距離を表す距離情報を生成する複数のステレオ画像処理部とを備える。本技術は、例えば、自動車に搭載できる。
摘要翻译：
此技术中，图像处理装置能够使用立体图像对包括由广角相机拍摄的广角图像，测量在宽的范围内的物体存在的距离，图像处理方法和车辆。本技术的图像处理装置中，根据通过将视野角在所述第一图像的成像时获得的投影角，所述多个基准平面图像由第一图像投影到多个平面上的虚拟球体根据通过将成像范围与第一图像重叠的第二图像成像时的视角分割而得到的每个投影角度，将虚拟球面上的多个第二图像通过投影平面上，所产生的多个参考平面图像和基准平面图像对应的图像对执行使用所述第二生成单元，用于生成所述多个基准平面图像立体图像处理以及多个立体图像处理单元，用于生成表示参考图像上到物体的距离的距离信息。本技术可以安装在例如汽车上。

10. 发明申请

WO2017189490A1 LIVE ACTION VOLUMETRIC VIDEO COMPRESSION / DECOMPRESSION AND PLAYBACK 审中-公开
标题翻译：实时动作音量视频压缩/减压和播放
公开(公告)号：WO2017189490A1
公开(公告)日：2017-11-02
申请号：PCT/US2017/029261
申请日：2017-04-25
申请人： HYPEVR
发明人： TRAN, Anthony , LAN, Jiang , JUANG, Jason , JIANG, Caoyang
IPC分类号： G01S17/02 , G01S17/89 , G06T7/60 , G06T19/20
CPC分类号： H04N13/161 , G01S17/023 , G06T7/40 , G06T7/593 , G06T7/60 , G06T19/006 , G06T2200/04 , H04N13/144 , H04N13/15 , H04N13/156 , H04N13/167 , H04N13/189 , H04N13/243 , H04N13/254 , H04N2013/0077 , H04N2013/0081 , H04N2013/0096 , H04N2213/002
摘要： A method for compressing geometric data and video is disclosed. The method includes receiving video and associated geometric data for a physical location, generating a background video from the video, and generating background geometric data for the geometric data outside of a predetermined distance from a capture point for the video as a skybox sphere at a non-parallax distance. The method further includes generating a geometric shape for a first detected object within the predetermined distance from the capture point from the geometric data, generating shape textures for the geometric shape from the video, and encoding the background video and shape textures as compressed video along with the geometric shape and the background geometric data as encoded volumetric video.
摘要翻译：公开了一种用于压缩几何数据和视频的方法。该方法包括：接收视频和相关联的几何数据的物理位置，生成从视频一个背景视频，并且在一个非生成从捕捉点的视频作为天空盒球预定距离的外侧的几何数据的背景几何数据 - 视差距离。该方法进一步包括从几何数据产生距离捕获点预定距离内的第一检测对象的几何形状，从视频生成几何形状的形状纹理，并且将背景视频和形状纹理作为压缩视频与几何形状和背景几何数据作为编码体积视频。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式