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

WO2014106651A1 EFFICIENT SCALABLE CODING CONCEPT 审中-公开
标题翻译：有效的可分级编码概念
公开(公告)号：WO2014106651A1
公开(公告)日：2014-07-10
申请号：PCT/EP2014/050065
申请日：2014-01-04
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： SÜHRING, Karsten , SCHIERL, Thomas , GEORGE, Valeri , MARPE, Detlev , SKUPIN, Robert
IPC分类号： H04N19/80 , H04N19/102 , H04N19/119 , H04N19/187 , H04N19/70 , H04N19/12 , H04N19/174 , H04N19/59 , H04N19/30 , H04N19/436 , H04N19/46 , H04N19/33
CPC分类号： H04N19/187 , H04N19/102 , H04N19/119 , H04N19/12 , H04N19/139 , H04N19/156 , H04N19/174 , H04N19/30 , H04N19/33 , H04N19/436 , H04N19/46 , H04N19/59 , H04N19/70 , H04N19/80
摘要： Scalable coding concepts such as for scalable video coding, are described. One aspect improves parallel decoding of inter-dependent layers of a multi-layer video data stream by introducing a long-term syntax element structure which, when assuming a certain value, guarantees to the video decoder that during a predetermined time period greater than the short-term syntax element's time intervals, the pictures of the dependent layer are subdivided so that borders between the spatial segments of the pictures of the second layer overlay every border of the spatial segments of the first layer. Another aspect concerns upsampling from base layer to enhancement layer. A syntax element is introduced which informs the decoder that the interpolation along the base layer's partitions is modified so as to not mix-up pixels/pels of neighboring partitions of the base layer picture. Another aspect introduces a long-term syntax element structure which allows the decoder to determine the inter-layer offset for a predetermined time period. Another aspect introduces a type indicator field which changes a way a layer indicator field within the NAL unit headers associated with the various layers is to be interpreted. Another aspect allows different codecs/standards to be used for the different layers. Another aspect inserts a syntax element structure into the multi-layer video data stream, which indicates the inter-layer offset for parallel decoding the pictures of base and enhancement layers in units of the base layer blocks.
摘要翻译：描述了诸如用于可缩放视频编码的可缩放编码概念。一个方面通过引入长期语法元素结构来改进多层视频数据流的相互依赖层的并行解码，该长期语法元素结构当假设某一值时保证视频解码器在大于短帧的预定时间段内 -term语法元素的时间间隔，依赖层的图像被细分，使得第二层的图片的空间段之间的边界覆盖第一层的空间段的每个边界。另一方面涉及从基层到增强层的上采样。引入语法元素，其通知解码器沿着基本层分区的内插被修改以便不混合基本层图像的相邻分区的像素/像素。另一方面引入了长期语法元素结构，其允许解码器在预定时间段内确定层间偏移。另一方面引入了类型指示符字段，其改变与各个层相关联的NAL单元头部内的层指示符字段将被解释的方式。另一方面允许不同的编解码器/标准被用于不同的层。另一方面将语法元素结构插入到多层视频数据流中，该多层视频数据流指示用于以基本层块为单位并行解码基础层和增强层的图像的层间偏移。

62. 发明申请

WO2014053514A1 SCALABLE VIDEO CODING USING BASE-LAYER HINTS FOR ENHANCEMENT LAYER MOTION PARAMETERS 审中-公开
标题翻译：使用基层智能的可扩展视频编码增强层运动参数
公开(公告)号：WO2014053514A1
公开(公告)日：2014-04-10
申请号：PCT/EP2013/070486
申请日：2013-10-01
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： HINZ, Tobias , LAKSHMAN, Haricharan , STEGEMANN, Jan , HELLE, Philipp , SIEKMANN, Mischa , SÜHRING, Karsten , MARPE, Detlev , SCHWARZ, Heiko , BARTNIK, Christian , KHAIRAT, Ali , KIRCHHOFFER, Heiner , WIEGAND, Thomas
IPC分类号： H04N7/36
CPC分类号： H04N19/119 , H04N19/105 , H04N19/107 , H04N19/11 , H04N19/117 , H04N19/137 , H04N19/157 , H04N19/176 , H04N19/18 , H04N19/187 , H04N19/198 , H04N19/33 , H04N19/463 , H04N19/48 , H04N19/503 , H04N19/587 , H04N19/59 , H04N19/593 , H04N19/61 , H04N19/615 , H04N19/70 , H04N19/82 , H04N19/865
摘要： Information available from coding/decoding the base layer, i.e. base-layer hints, is exploited to render the motion-compensated prediction of the enhancement layer more efficient by more efficiently coding the enhancement layer motion parameters.
摘要翻译：利用从基层编码/解码获得的信息，即基层提示，通过对增强层运动参数进行更有效的编码，使增强层的运动补偿预测更有效率。

63. 发明申请

WO2014001573A1 VIDEO DATA STREAM CONCEPT 审中-公开
标题翻译：视频数据流概念
公开(公告)号：WO2014001573A1
公开(公告)日：2014-01-03
申请号：PCT/EP2013/063853
申请日：2013-07-01
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： SCHIERL, Thomas , GEORGE, Valeri , HENKEL, Anastasia , MARPE, Detlev , GRÜNEBERG, Karsten , SKUPIN, Robert
IPC分类号： H04N7/26 , H04N21/00
CPC分类号： H04N19/68 , G06F15/173 , H04L12/56 , H04L12/66 , H04L47/10 , H04L47/31 , H04N19/167 , H04N19/174 , H04N19/188 , H04N19/423 , H04N19/436 , H04N19/46 , H04N19/55 , H04N19/67 , H04N19/70 , H04N19/91 , H04N21/234327 , H04N21/4621 , H04N21/4728
摘要： Decoder retrieval timing information. ROI information and tile identification information are conveyed within a video data stream at a level which allows for an easy access by network entities such as MANEs or decoder. In order to reach such a level, information of such types are conveyed within a video data stream by way of packets interspersed into packets of access units of a video data stream. In accordance with an embodiment, the interspersed packets are of a removable packet type, i.e. the removal of these interspersed packets maintains the decoder's ability to completely recover the video content conveyed via the video data stream.
摘要翻译：解码器检索定时信息。 ROI信息和瓦片识别信息以视频数据流的形式传送，允许诸如MANE或解码器之类的网络实体容易地访问。为了达到这样的水平，通过分组到视频数据流的访问单元的分组中的分组，在视频数据流内传送这种类型的信息。根据实施例，散布的分组是可移动分组类型，即，去除这些分散的分组保持解码器完全恢复经由视频数据流传送的视频内容的能力。

64. 发明申请

WO2013153227A2 SCALABLE DATA STREAM AND NETWORK ENTITY 审中-公开
标题翻译：可扩展数据流和网络实体
公开(公告)号：WO2013153227A2
公开(公告)日：2013-10-17
申请号：PCT/EP2013/057803
申请日：2013-04-15
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： SCHIERL, Thomas , GEORGE, Valeri , HENKEL, Anastasia , MARPE, Detlev , GRÜNEBERG, Karsten , SKUPIN, Robert
IPC分类号： H04N7/26
CPC分类号： H04N19/00884 , H04N19/13 , H04N19/174 , H04N19/188 , H04N19/436 , H04N19/46 , H04N19/463 , H04N19/593 , H04N19/70 , H04N19/91
摘要： The handling of scalable data streams by network entities is rendered less complex by, in the data stream, accompanying packets which actually carry data by packets of a different packet type which have a scalability axes descriptor defining the number of scalability axes and a semantic meaning thereof. By this measure, it is sufficient if the data carrying packets comprise an operation point identifier identifying an operation point associated with a respective packet within the scalability space spanned by the scalability axis. By this measure, the packet operation point identifiers provide enough information to a network entity so as to associate these packets with their operation point in the scalability space provided the network entity has access to this scalability axes descriptor provided by the packet of the predetermined packet type. In another aspect, the handling of scalable data streams by network entities is rendered more efficient by conveying level and/or profile descriptors using packets other than the data carrying packets, with profile and/or level descriptors being divided up into a first set explicitly signaling the definition of the available coding option set and/or available syntax element value range for a respective operation point, and a second set of profile and/or level descriptors which signal the definition of the available coding option set and/or available syntax element value range for their respective operation points by reference another profile and/or level descriptor.
摘要翻译：通过网络实体对可伸缩数据流的处理在数据流中由于具有定义可伸缩性轴的数量的可伸缩性轴描述符的实际携带数据的不同分组类型的数据包以及其语义意义而使数据流变得不复杂。通过这种措施，如果携带数据的数据包包括识别与可扩展性轴跨越的可扩展性空间内的相应分组相关联的操作点的操作点标识符就足够了。通过这种措施，分组操作点标识符向网络实体提供足够的信息，以便在可扩展性空间中将这些分组与其操作点相关联，只要网络实体可以访问由预定分组类型的分组提供的该可伸缩性轴描述符。在另一方面，通过使用不同于携带分组的数据的分组传送级别和/或简档描述符，通过网络实体来处理可伸缩数据流变得更有效率，其中简档和/或级别描述符被划分成第一组显式信令可用编码选项集的定义和/或相应操作点的可用语法元素值范围，以及第二组简档和/或级别描述符，其指示可用编码选项集的定义和/或可用语法元素值通过参考另一个配置文件和/或电平描述符为其各自的操作点的范围。

65. 发明申请

WO2013068564A1 EFFECTIVE WEDGELET PARTITION CODING USING SPATIAL PREDICTION 审中-公开
标题翻译：使用空间预测的有效的卫星分割编码
公开(公告)号：WO2013068564A1
公开(公告)日：2013-05-16
申请号：PCT/EP2012/072326
申请日：2012-11-09
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： MERKLE, Philipp , BARTNIK, Christian , LAKSHMAN, Haricharan , MARPE, Detlev , MÜLLER, Karsten , WIEGAND, Thomas , TECH, Gerhard
IPC分类号： H04N7/34 , H04N7/26
CPC分类号： H04N19/51 , H04N19/105 , H04N19/119 , H04N19/14 , H04N19/157 , H04N19/176 , H04N19/196 , H04N19/46 , H04N19/52 , H04N19/593 , H04N19/597 , H04N19/70 , H04N19/82 , H04N19/96
摘要： In accordance with a first aspect, the intra prediction direction of a neighboring, intra- predicted block is used in order to predict the extension direction of the wedgelet separation line of a current block, thereby reducing the side information rate necessary in order to convey the partitioning information. In accordance with a second aspect, the idea is that previously reconstructed samples, i.e. reconstructed values of blocks preceding the current block in accordance with the coding/decoding order allow for at least a prediction of a correct placement of a starting point of the wedgelet separation line, namely by placing the starting point of the wedgelet separation line at a position of a maximum change between consecutive ones of a sequence of reconstructed values of samples of a line of samples extending adjacent to the current block along a circumference thereof. Both aspects may be used individually or in combination.
摘要翻译：根据第一方面，使用相邻的帧内预测块的帧内预测方向来预测当前块的楔形分离线的延伸方向，从而减少为了传送所述帧预测块所需的侧信息速率分区信息。根据第二方面，该想法是，先前重建的样本，即根据编码/解码顺序的当前块之前的块的重建值，允许至少预测楔形分离的起始点的正确放置即通过将楔形分离线的起始点放置在沿着其圆周与当前块相邻延伸的一系列样本的样本重建值的序列中的连续的重复值之间的最大变化的位置。这两个方面可以单独使用或组合使用。

66. 发明申请

WO2021209470A1 DECODER FOR DECODING WEIGHT PARAMETERS OF A NEURAL NETWORK, ENCODER, METHODS AND ENCODED REPRESENTATION USING PROBABILITY ESTIMATION PARAMETERS 审中-公开
公开(公告)号：WO2021209470A1
公开(公告)日：2021-10-21
申请号：PCT/EP2021/059594
申请日：2021-04-13
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： HAASE, Paul , MÜLLER, Karsten , KIRCHHOFFER, Heiner , SCHWARZ, Heiko , MARPE, Detlev , WIEGAND, Thomas
IPC分类号： H03M7/30 , H03M7/40 , G06N3/063
摘要： Embodiments according to the invention comprise a decoder for decoding weight Parameters of a neural network, wherein the decoder is configured to obtain a plurality of neural network Parameters, e.g., at least one of entries w, of matrix W, b, μ, σ2, σ, γ, and/or β, of the neural network on the basis of an encoded bitstream. Furthermore, the decoder is configured to decode the neural network Parameters of the neural network, e.g., a quantized Version of the neural network Parameters, using a context-dependent arithmetic decoding, e.g., using a context-adaptive binary arithmetic decoding (CABAC). Optionally, probabilities of bin values may be determined for different contexts, wherein, for example, each bin is associated with a context. Moreover, the decoder is configured to obtain a probability estimate, which may, for example, be associated with a context, for a, e.g. arithmetic, decoding of a bin of a number representation of a neural network parameter, e.g. on the basis of one or more previously decoded neural network Parameters or bins thereof, using one or more probability estimation Parameters. In addition, the decoder is configured to use different probability estimation Parameter values for a decoding of different neural network Parameters and/or to use different probability estimation parameter values for a decoding of bins associated with different context models. Further embodiments comprise a decoder configured to use different probability estimation parameter values for a decoding of neural network Parameters associated with different layers of the neural network. Corresponding encoders, methods and encoded representations are also disclosed.

67. 发明申请

WO2021123438A1 CONCEPTS FOR CODING NEURAL NETWORKS PARAMETERS 审中-公开
公开(公告)号：WO2021123438A1
公开(公告)日：2021-06-24
申请号：PCT/EP2020/087489
申请日：2020-12-21
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： HAASE, Paul , KIRCHHOFFER, Heiner , SCHWARZ, Heiko , MARPE, Detlev , WIEGAND, Thomas
IPC分类号： G06N3/02
摘要： Embodiments according to a first aspect of the present invention are based on the idea, that neural network parameters may be compressed more efficiently by using a non-constant quantizer, but varying same during coding the neural network parameters, namely by selecting a set of reconstruction levels depending on quantization indices decoded from, or respectively encoded, into the data stream for previous or respectively previously encoded neural network parameters. Embodiments according to a second aspect of the present invention are based on the idea that a more efficient neural network coding may be achieved when done in stages – called reconstruction layers to distinguish them from the layered composition of the neural network in neural layers – and if the parametrizations provided in these stages are then, neural network parameter-wise combined to yield a neural network parametrization improved compared to any of the stages.

68. 发明申请

WO2021110943A1 MATRIX BASED INTRA PREDICTION WITH MODE-GLOBAL SETTINGS 审中-公开
公开(公告)号：WO2021110943A1
公开(公告)日：2021-06-10
申请号：PCT/EP2020/084691
申请日：2020-12-04
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： PFAFF, Jonathan , HINZ, Tobias , HELLE, Philipp , MERKLE, Philipp , STALLENBERGER, Björn , SCHÄFER, Michael , BROSS, Benjamin , WINKEN, Martin , SIEKMANN, Mischa , SCHWARZ, Heiko , MARPE, Detlev , WIEGAND, Thomas
IPC分类号： H04N19/11 , H04N19/147 , H04N19/176 , H04N19/593 , H04N19/70 , H04N19/156
摘要： Apparatus for decoding a predetermined block (18) of a picture using intra-prediction, configured to read, from the data stream (12), a mode index (200), the mode index pointing to one out of a list (204) of matrix-based intra-prediction modes. Additionally, the apparatus is configured to predict samples (108) of the predetermined block (18) by computing a matrix- vector product (206) between an input vector (102) derived from reference samples (17) in a neighborhood of the predetermined block (18) and a prediction matrix (19) associated with the matrix-based intra-prediction mode (k) pointed to by the mode index (200) and associating components (210) of an output vector (208) obtained by the matrix-vector product (206) onto sample positions (104) of the predetermined block. For each matrix-based intra-prediction mode, all entries of the prediction matrix (19) associated with the respective matrix-based intra-prediction mode are represented by a fixed point representation of a predetermined bit-depth, the predetermined bit-depth being equal for the matrix-based intra- prediction modes. Furthermore, the apparatus is configured to, for each matrix-based intra-prediction mode, compute the matrix-vector product (206) between the input vector (102) and the prediction matrix (19) associated with the respective matrix-based intra-prediction mode (k) by performing, for each component of the output vector, a right shift at a number of bits which is equal for the matrix-based intra-prediction modes.

69. 发明申请

WO2021105255A1 CODING CONCEPTS FOR A TRANSFORMED REPRESENTATION OF A SAMPLE BLOCK 审中-公开
公开(公告)号：WO2021105255A1
公开(公告)日：2021-06-03
申请号：PCT/EP2020/083440
申请日：2020-11-25
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： BROSS, Benjamin , DE LUXÁN HERNÁNDEZ, Santiago , SCHWARZ, Heiko , MARPE, Detlev , WIEGAND, Thomas
IPC分类号： H04N19/129 , H04N19/136 , H04N19/176 , H04N19/13 , H04N19/12 , H04N19/70
摘要： Decoder for decoding a transformed representation of a sample block from a data stream. If a first coded coefficient is located inside a predetermined subarea of the transform coefficient block and if the underlying transform is within a first set of available transforms, the decoder decodes coefficients along a first coefficient scan order. If the transform is within a second set of available transforms, the decoder decodes coefficients located within the predetermined subarea along a second coefficient scan order, and infers that coefficients located outside the predetermined subarea are zero. The first coefficient scan order is so that coefficients outside the predetermined subarea are scanned between two transform coefficients located inside the predetermined subarea. The second coefficient scan order does not scan any coefficient outside the predetermined subarea between scanning the coefficients within the predetermined subarea.

70. 发明申请

WO2021064013A2 NEURAL NETWORK REPRESENTATION FORMATS 审中-公开
公开(公告)号：WO2021064013A2
公开(公告)日：2021-04-08
申请号：PCT/EP2020/077352
申请日：2020-09-30
申请人： FRAUNHOFER-GESELLSCHAFT ZUR FÖRDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
发明人： MATLAGE, Stefan , HAASE, Paul , KIRCHHOFFER, Heiner , MÜLLER, Karsten , SAMEK, Wojciech , WIEDEMANN, Simon , MARPE, Detlev , SCHIERL, Thomas , SÁNCHEZ DE LA FUENTE, Yago , SKUPIN, Robert , WIEGAND, Thomas
IPC分类号： G06N3/04 , G06N3/08 , G06N3/10 , H03M7/00 , G06N3/0454 , G06N3/0481 , G06N3/084 , G06N3/105 , H03M7/4018 , H03M7/6023 , H03M7/70
摘要： Data stream (45) having a representation of a neural network (10) encoded thereinto, the data stream (45) comprising serialization parameter (102) indicating a coding order (104) at which neural network parameters (32), which define neuron interconnections (22, 24) of the neural network (10), are encoded into the data stream (45).

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