专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

61. 发明申请

US20220076117A1 METHODS FOR GENERATING A DEEP NEURAL NET AND FOR LOCALISING AN OBJECT IN AN INPUT IMAGE, DEEP NEURAL NET, COMPUTER PROGRAM PRODUCT, AND COMPUTER-READABLE STORAGE MEDIUM 有权
公开(公告)号：US20220076117A1
公开(公告)日：2022-03-10
申请号：US17279087
申请日：2019-08-28
申请人： Peter Amon , Sanjukta Ghosh , Andreas Hutter
发明人： Peter Amon , Sanjukta Ghosh , Andreas Hutter
IPC分类号： G06N3/08 , G06T7/10 , G06T7/70
摘要： Methods for generating a deep neural net and for localizing an object in an input image, the deep neural net, a corresponding computer program product, and a corresponding computer-readable storage medium are provided. A discriminative counting model is trained to classify images according to a number of objects of a predetermined type depicted in each of the images, and a segmentation model is trained to segment images by classifying each pixel according to what image part the pixel belongs to. Parts and/or features of both models are combined to form the deep neural net. The deep neural net is adapted to generate, in a single forward pass, a map indicating locations of any objects for each input image.

62. 发明授权

US09615087B2 Method for processing a compressed video stream 有权
公开(公告)号：US09615087B2
公开(公告)日：2017-04-04
申请号：US13980996
申请日：2012-01-16
申请人： Peter Amon , Andreas Hutter , André Kaup , Marcus Laumer
发明人： Peter Amon , Andreas Hutter , André Kaup , Marcus Laumer
IPC分类号： H04N9/80 , H04N19/48 , H04N5/14 , H04N19/172 , H04N19/146 , H04N19/87 , H04N19/169
CPC分类号： H04N19/48 , H04N5/147 , H04N19/146 , H04N19/172 , H04N19/188 , H04N19/87
摘要： For a compressed video stream formed of a sequence of compressed data packets, each compressed packet is assigned to a frame by processing the undecoded data packet. A frame includes packets. A first plurality of frames is assigned to a first sliding window and a second plurality of frames is assigned to a second sliding window. The first and second sliding windows have different frames. A first window indicator is calculated by determining a first indicator representing an average number of packets per frame within the first window, and/or a second indicator representing size information of the data within the first window. A second window indicator is calculated by determining a third indicator representing an average number of packets per frame within the second window, and/or a fourth indicator representing size information of the data within the second window. The first and the second window indicators are compared.

63. 发明申请

US20150227792A1 Methods and Devices for Object Detection 审中-公开
标题翻译：对象检测的方法和设备
公开(公告)号：US20150227792A1
公开(公告)日：2015-08-13
申请号：US14177126
申请日：2014-02-10
申请人： Peter Amon , Jan Ernst , Andreas Hutter , Johannes Rehm , Vivek Kumar Singh
发明人： Peter Amon , Jan Ernst , Andreas Hutter , Johannes Rehm , Vivek Kumar Singh
IPC分类号： G06K9/00 , G06K9/62 , G06K9/46
CPC分类号： G06K9/4671
摘要： Object detection includes providing an image and determining at least one feature point. The at least one feature point defines a location of an image patch used for determining a feature descriptor, and the image patch defines an image area of the image. The feature descriptor is generated based on respective image intensities of a number of respective pairs of pixels with two dimensional coordinates located inside the image patch. An n-th component of the feature descriptor for an n-th pair of pixels is derived. A threshold is set depending on the number. The feature descriptor is generated by an arrangement of the M components. An indication signal is generated for a detected object when the feature descriptor is within a predefined distance to a reference feature descriptor.
摘要翻译：对象检测包括提供图像和确定至少一个特征点。所述至少一个特征点定义用于确定特征描述符的图像块的位置，并且所述图像块定义所述图像的图像区域。特征描述符是基于位于图像块内的具有二维坐标的多个像素对的数量的各自的图像强度来生成的。导出第n对像素的特征描述符的第n个分量。根据数量设置阈值。特征描述符由M个分量的排列产生。当特征描述符处于与参考特征描述符的预定距离内时，为检测到的对象生成指示信号。

64. 发明申请

US20140003517A1 COMPRESSION AND DECOMPRESSION OF REFERENCE IMAGES IN A VIDEO CODING DEVICE 有权
标题翻译：在视频编码设备中的参考图像的压缩和分解
公开(公告)号：US20140003517A1
公开(公告)日：2014-01-02
申请号：US13979601
申请日：2012-01-12
申请人： Bernhard Agthe , Peter Amon , Gero Bäse , Andreas Hutter , Robert Kutka , Norbert Oertel
发明人： Bernhard Agthe , Peter Amon , Gero Bäse , Andreas Hutter , Robert Kutka , Norbert Oertel
IPC分类号： H04N7/26
CPC分类号： H04N19/423 , H04N19/117 , H04N19/124 , H04N19/147 , H04N19/159 , H04N19/428 , H04N19/46
摘要： A reconstructed image is provided within the framework of predictive coding. The reconstructed image is compressed according to a coding mode of an image coding device and stored in a reference image memory.
摘要翻译：在预测编码的框架内提供重建的图像。重建图像根据图像编码装置的编码模式被压缩并存储在参考图像存储器中。

65. 发明授权

US08428140B2 Coding and decoding method and device 有权
标题翻译：编码和解码方法及装置
公开(公告)号：US08428140B2
公开(公告)日：2013-04-23
申请号：US11659211
申请日：2005-07-27
申请人： Peter Amon , Andreas Hutter , Benoit Timmermann
发明人： Peter Amon , Andreas Hutter , Benoit Timmermann
IPC分类号： H04N7/12
CPC分类号： H04N19/59 , H04N19/119 , H04N19/147 , H04N19/172 , H04N19/176 , H04N19/187 , H04N19/33 , H04N19/36 , H04N19/517 , H04N19/61 , H04N19/70 , H04N19/96
摘要： In a method for the video coding of image sequences images in the image sequence are coded in a scaled manner, in such a way that the video data produced contains information which permits the images to be represented in a plurality of differing stages of image resolution, the latter being defined by the number of pixels per image representation. The coding is block-based, in such a way that to describe a displacement of parts of one of the images, said displacement being contained in the image sequence, at least one block structure that describes the displacement is created. Said block structure is configured from one block, which is subdivided into sub-blocks, whereby some of the sub-blocks are further subdivided into successively smaller sub-blocks. A first block structure is temporarily created for at least one first resolution stage and a second block structure is created for a second resolution stage, the first resolution stage having a lower number of pixels than the second resolution stage. In addition, the second block structure is compared with the first block structure to determine the differences in structure, in order to create a modified second block structure on the basis of characteristics of the structural differences, said modified block structure representing part of the second block structure and forming the basis for the coding of the image sequence.
摘要翻译：在用于图像序列的视频编码的方法中，以缩放的方式对图像序列中的图像进行编码，使得所生成的视频数据包含允许以多个不同的图像分辨率表示图像的信息，后者由每个图像表示的像素数量定义。编码是基于块的，以这样的方式，为了描述图像之一的部分的位移，所述位移包含在图像序列中，形成描述位移的至少一个块结构。所述块结构由一个块构成，其被细分为子块，由此一些子块被进一步细分为相继较小的子块。暂时为至少一个第一分辨率阶段创建第一块结构，并为第二分辨率阶段创建第二块结构，第一分辨率阶段具有比第二分辨率阶段少的像素数。此外，将第二块结构与第一块结构进行比较以确定结构的差异，以便基于结构差异的特征创建修改的第二块结构，所述修改的块结构表示第二块的一部分结构和形成图像序列编码的基础。

66. 发明授权

US08108539B2 Method for synchronising scene data files and media data flows in an unindirectional broadcast system 有权
标题翻译：用于在非定向广播系统中同步场景数据文件和媒体数据流的方法
公开(公告)号：US08108539B2
公开(公告)日：2012-01-31
申请号：US12312230
申请日：2007-10-26
申请人： Jörg Heuer , Andreas Hutter
发明人： Jörg Heuer , Andreas Hutter
IPC分类号： G06F15/16
CPC分类号： H04N21/23617 , H04N21/2368 , H04N21/242 , H04N21/4341
摘要： A method for the temporal synchronization of data packets of at least one media data flow with scene data files describing at least one scene and/or the data packets of said data files, in a unidirectional broadcast system. The data packets of the media data flow are each provided with a first time mark relating to a first reference time, and the scene data files and/or the data packets thereof are each provided with a second time mark relating to a second reference time. At least one time mark allocation data flow is transmitted from the transmitter to the receiver, the first time marks of the media data flow and the second time marks of the scene data files and/or the data packets thereof being respectively associated with a third time mark relating to a third reference time.
摘要翻译：一种用于在单向广播系统中至少一个媒体数据流的数据分组与描述所述数据文件的至少一个场景和/或数据分组的场景数据文件的时间同步的方法。媒体数据流的数据分组各自被提供有与第一参考时间相关的第一时间标记，并且场景数据文件和/或其数据分组被提供有与第二参考时间有关的第二时间标记。至少一个时间标记分配数据流从发射机发送到接收机，媒体数据流的第一时间标记和场景数据文件的第二时间标记和/或其数据分组分别与第三时间相关联与第三个参考时间相关的标记。

67. 发明授权

US08024366B2 System and method for the improved encoding/decoding of binary representations of structured, documents 有权
标题翻译：用于改进结构化文档的二进制表示的编码/解码的系统和方法
公开(公告)号：US08024366B2
公开(公告)日：2011-09-20
申请号：US10482518
申请日：2002-06-25
申请人： Jörg Heuer , Andreas Hutter , Ulrich Niedermeier
发明人： Jörg Heuer , Andreas Hutter , Ulrich Niedermeier
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： H04N21/2353 , H04N21/235 , H04N21/435
摘要： A schema definition, which is known to the encoder and to the decoder or transmitted from the encoder to the decoder, specifies information for calculating code concerning the elements or data types contained in the name subspace and enabling this information to be clearly assigned in this manner to each code in the name subspace. Only a portion of the entire name space including the number of the elements or data types lacking for the code calculation have to be transmitted for the configuration of a code which generally means a significantly smaller volume of data than the volume of data that would be required for the entire name space. In addition, the code tables are smaller than in entirely known name spaces which requires less storage space in the codec and enables a faster encoding and decoding.
摘要翻译：编码器和解码器已知的或从编码器传送到解码器的模式定义指定用于计算与名称子空间中包含的元素或数据类型有关的代码的信息，并且以这种方式清楚地分配该信息到名称子空间中的每个代码。必须发送整个名称空间的一部分，包括缺少代码计算的元素或数据类型的数量，以配置代码，这通常意味着数据量要比所需数据量小得多为整个名称空间。此外，代码表小于完全已知的名称空间，这在编解码器中需要较少的存储空间，并能够实现更快的编码和解码。

68. 发明授权

US07797346B2 Method for improving the functionality of the binary representation of MPEG-7 and other XML based content descriptions 有权
标题翻译：改进MPEG-7和其他基于XML的内容描述的二进制表示的功能的方法
公开(公告)号：US07797346B2
公开(公告)日：2010-09-14
申请号：US10467047
申请日：2002-01-31
申请人： Jörg Heuer , Andreas Hutter , Ulrich Niedermeier
发明人： Jörg Heuer , Andreas Hutter , Ulrich Niedermeier
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： H04N21/2353 , G06F17/30244 , G06F17/3061 , H04N21/235 , H04N21/435
摘要： A method is provided for encoding string-based data types, which have, for example, a tree-type structure, in such a way that signal bits (flags) for optional sections and/or for length information concerning the codes that are of variable length are placed in a defined sequence at the beginning of the created bitstream, thus accelerating the search for data via a sorting process and bit mask queries. A more efficient encoding is also provided of the #position field by, among other things, the assignment of code fragments for each group of regular expressions.
摘要翻译：提供了一种用于编码基于字符串的数据类型的方法，该数据类型具有例如树型结构，使得可选部分的信号位（标志）和/或关于可变部分的代码的长度信息长度以所创建的比特流的开始处以限定的顺序放置，从而通过排序处理和位掩码查询加速对数据的搜索。除了别的以外，还可以通过#position字段为每个正则表达式组分配代码片段来提供更有效的编码。

69. 发明申请

US20100049863A1 METHOD FOR SYNCHRONISING SCENE DATA FILES AND MEDIA DATA FLOWS IN AN UNIDIRECTIONAL BROADCAST SYSTEM 有权
标题翻译：同步广播系统中场景数据文件和媒体数据流的同步方法
公开(公告)号：US20100049863A1
公开(公告)日：2010-02-25
申请号：US12312230
申请日：2007-10-26
申请人： Jörg Heuer , Andreas Hutter
发明人： Jörg Heuer , Andreas Hutter
IPC分类号： G06F15/16
CPC分类号： H04N21/23617 , H04N21/2368 , H04N21/242 , H04N21/4341
摘要： A method for the temporal synchronization of data packets of at least one media data flow with scene data files describing at least one scene and/or the data packets of said data files, in a unidirectional broadcast system. The data packets of the media data flow are each provided with a first time mark relating to a first reference time, and the scene data files and/or the data packets thereof are each provided with a second time mark relating to a second reference time. At least one time mark allocation data flow is transmitted from the transmitter to the receiver, the first time marks of the media data flow and the second time marks of the scene data files and/or the data packets thereof being respectively associated with a third time mark relating to a third reference time.
摘要翻译：一种用于在单向广播系统中至少一个媒体数据流的数据分组与描述所述数据文件的至少一个场景和/或数据分组的场景数据文件的时间同步的方法。媒体数据流的数据分组各自被提供有与第一参考时间相关的第一时间标记，并且场景数据文件和/或其数据分组被提供有与第二参考时间有关的第二时间标记。至少一个时间标记分配数据流从发射机发送到接收机，媒体数据流的第一时间标记和场景数据文件的第二时间标记和/或其数据分组分别与第三时间相关联与第三个参考时间相关的标记。

70. 发明授权

US07627586B2 Method for encoding a structured document 有权
标题翻译：编码结构化文档的方法
公开(公告)号：US07627586B2
公开(公告)日：2009-12-01
申请号：US10548262
申请日：2004-02-27
申请人： Jörg Heuer , Andreas Hutter , Andrea Kofler-Vogt
发明人： Jörg Heuer , Andreas Hutter , Andrea Kofler-Vogt
IPC分类号： G06F7/00
CPC分类号： G06F17/2258 , G06F17/2247 , H04N21/235 , H04N21/2353 , H04N21/435 , Y10S707/99942
摘要： A method is for encoding a structured document, particularly an XML document, during which the contents of the document are converted into a binary representation. This binary representation is divided into encoding units, which form an encoded data flow and can be read out from the encoded data flow. The encoded data flow contains configuration data, with which configuration information concerning the division of the binary representation into encoding units can be read out before the reading out of one of more encoding units.
摘要翻译：一种用于编码结构化文档，特别是XML文档的方法，在该文档期间，文档的内容被转换成二进制表示。该二进制表示被分成编码单元，其形成编码数据流并且可以从编码数据流中读出。编码数据流包含配置数据，在读出多个编码单元之一之前，可以读出关于将二进制表示划分成编码单元的配置信息。

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