公开(公告)号：US20130114703A1

公开(公告)日：2013-05-09

申请号：US13725980

申请日：2012-12-21

申请人： Euclid Discoveries, LLC

发明人： Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Charles P. Pace

IPC分类号： H04N7/26 ,  H04N7/32

CPC分类号： H04N19/20 ,  H04N19/503

摘要： A model-based compression codec applies higher-level modeling to produce better predictions than can be found through conventional block-based motion estimation and compensation. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks and related to specific blocks of video data to be encoded. The tracking information is used to produce model-based predictions for those blocks of data, enabling more efficient navigation of the prediction search space than is typically achievable through conventional motion estimation methods. A hybrid framework enables modeling of data at multiple fidelities and selects the appropriate level of modeling for each portion of video data.

摘要翻译： 基于模型的压缩编解码器应用较高级别的建模以产生比通过传统的基于块的运动估计和补偿可以发现的更好的预测。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道并与要编码的视频数据的特定块有关。 跟踪信息用于为这些数据块产生基于模型的预测，使得能够比通常通过常规运动估计方法可实现的预测搜索空间更有效地导航。 混合框架可以对多个保真度的数据进行建模，并为视频数据的每个部分选择适当的建模级别。

公开(公告)号：US20150124874A1

公开(公告)日：2015-05-07

申请号：US14527477

申请日：2014-10-29

申请人： EUCLID DISCOVERIES, LLC

发明人： Charles P. Pace ,  Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Richard Wingard

IPC分类号： H04N19/17 ,  H04N19/136

CPC分类号： H04N19/54 ,  H04N19/167

摘要： Systems and methods of improving video encoding/decoding efficiency may be provided. A feature-based processing stream is applied to video data having a series of video frames. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks, and each track is given a representative, characteristic feature. Similar characteristic features are clustered and then stored in a model library, for reuse in the compression of other videos. A model-based compression framework makes use of the preserved model data by detecting features in a new video to be encoded, relating those features to specific blocks of data, and accessing similar model information from the model library. The formation of model libraries can be specialized to include personal, “smart” model libraries, differential libraries, and predictive libraries. Predictive model libraries can be modified to handle a variety of demand scenarios.

摘要翻译： 可以提供改善视频编码/解码效率的系统和方法。 基于特征的处理流被应用于具有一系列视频帧的视频数据。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道，并且每个轨道被赋予代表性的特征。 类似的特征特征被聚类，然后存储在模型库中，以便在压缩其他视频中重用。 基于模型的压缩框架通过检测要编码的新视频中的特征来使用保留的模型数据，将这些特征与特定的数据块相关联，以及从模型库访问相似的模型信息。 模型库的形成可以专门包括个人，“智能”模型库，差异库和预测库。 可以修改预测模型库来处理各种需求情况。

公开(公告)号：US20130107948A1

公开(公告)日：2013-05-02

申请号：US13725940

申请日：2012-12-21

申请人： Euclid Discoveries, LLC

发明人： Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Charles P. Pace

IPC分类号： H04N7/26

CPC分类号： H04N19/20 ,  H04N19/503

摘要： A model-based compression codec applies higher-level modeling to produce better predictions than can be found through conventional block-based motion estimation and compensation. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks and related to specific blocks of video data to be encoded. The tracking information is used to produce model-based predictions for those blocks of data, enabling more efficient navigation of the prediction search space than is typically achievable through conventional motion estimation methods. A hybrid framework enables modeling of data at multiple fidelities and selects the appropriate level of modeling for each portion of video data.

公开(公告)号：US09743078B2

公开(公告)日：2017-08-22

申请号：US13797644

申请日：2013-03-12

申请人： Euclid Discoveries, LLC

发明人： Darin DeForest ,  Charles P. Pace ,  Nigel Lee ,  Renato Pizzorni

IPC分类号： H04N7/26 ,  H04N19/50 ,  H04N19/543

CPC分类号： H04N19/50 ,  H04N19/23 ,  H04N19/51 ,  H04N19/543 ,  H04N19/85

摘要： A model-based compression codec applies higher-level modeling to produce better predictions than can be found through conventional block-based motion estimation and compensation. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks and related to specific blocks of video data to be encoded. The tracking information is used to produce model-based predictions for those blocks of data, enabling more efficient navigation of the prediction search space than is typically achievable through conventional motion estimation methods. A hybrid framework enables modeling of data at multiple fidelities and selects the appropriate level of modeling for each portion of video data. A compliant-stream version of the model-based compression codec uses the modeling information indirectly to improve compression while producing bitstreams that can be interpreted by standard decoders.

公开(公告)号：US09578345B2

公开(公告)日：2017-02-21

申请号：US13725940

申请日：2012-12-21

申请人： Euclid Discoveries, LLC

发明人： Darin DeForest ,  Charles P. Pace ,  Nigel Lee ,  Renato Pizzorni

IPC分类号： H04N19/20 ,  H04N19/503

CPC分类号： H04N19/20 ,  H04N19/503

摘要： A model-based compression codec applies higher-level modeling to produce better predictions than can be found through conventional block-based motion estimation and compensation. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks and related to specific blocks of video data to be encoded. The tracking information is used to produce model-based predictions for those blocks of data, enabling more efficient navigation of the prediction search space than is typically achievable through conventional motion estimation methods. A hybrid framework enables modeling of data at multiple fidelities and selects the appropriate level of modeling for each portion of video data.

摘要翻译： 基于模型的压缩编解码器应用较高级别的建模以产生比通过传统的基于块的运动估计和补偿可以发现的更好的预测。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道并与要编码的视频数据的特定块有关。 跟踪信息用于为这些数据块产生基于模型的预测，使得能够比通常通过常规运动估计方法可实现的预测搜索空间更有效地导航。 混合框架可以对多个保真度的数据进行建模，并为视频数据的每个部分选择适当的建模级别。

公开(公告)号：US09532069B2

公开(公告)日：2016-12-27

申请号：US14527477

申请日：2014-10-29

申请人： EUCLID DISCOVERIES, LLC

发明人： Charles P. Pace ,  Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Richard Wingard

IPC分类号： H04N19/54 ,  H04N19/167

CPC分类号： H04N19/54 ,  H04N19/167

摘要： Systems and methods of improving video encoding/decoding efficiency may be provided. A feature-based processing stream is applied to video data having a series of video frames. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks, and each track is given a representative, characteristic feature. Similar characteristic features are clustered and then stored in a model library, for reuse in the compression of other videos. A model-based compression framework makes use of the preserved model data by detecting features in a new video to be encoded, relating those features to specific blocks of data, and accessing similar model information from the model library. The formation of model libraries can be specialized to include personal, “smart” model libraries, differential libraries, and predictive libraries. Predictive model libraries can be modified to handle a variety of demand scenarios.

摘要翻译： 可以提供改善视频编码/解码效率的系统和方法。 基于特征的处理流被应用于具有一系列视频帧的视频数据。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道，并且每个轨道被给予代表性的特征。 类似的特征特征被聚类，然后存储在模型库中，以便在压缩其他视频中重用。 基于模型的压缩框架通过检测要编码的新视频中的特征来使用保留的模型数据，将这些特征与特定的数据块相关联，以及从模型库访问相似的模型信息。 模型库的形成可以专门包括个人，“智能”模型库，差异库和预测库。 可以修改预测模型库来处理各种需求情况。

公开(公告)号：US08902971B2

公开(公告)日：2014-12-02

申请号：US13772230

申请日：2013-02-20

申请人： Euclid Discoveries, LLC

发明人： Charles P. Pace ,  Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Richard Wingard

IPC分类号： H04N7/26 ,  H04N19/119 ,  G06T9/00

CPC分类号： H04N19/149 ,  G06T9/001

摘要： Systems and methods of improving video encoding/decoding efficiency may be provided. A feature-based processing stream is applied to video data having a series of video frames. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks, and each track is given a representative, characteristic feature. Similar characteristic features are clustered and then stored in a model library, for reuse in the compression of other videos. A model-based compression framework makes use of the preserved model data by detecting features in a new video to be encoded, relating those features to specific blocks of data, and accessing similar model information from the model library. The formation of model libraries can be specialized to include personal, “smart” model libraries, differential libraries, and predictive libraries. Predictive model libraries can be modified to handle a variety of demand scenarios.

摘要翻译： 可以提供改善视频编码/解码效率的系统和方法。 基于特征的处理流被应用于具有一系列视频帧的视频数据。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道，并且每个轨道被给予代表性的特征。 类似的特征特征被聚类，然后存储在模型库中，以便在压缩其他视频中重用。 基于模型的压缩框架通过检测要编码的新视频中的特征来使用保留的模型数据，将这些特征与特定的数据块相关联，以及从模型库访问相似的模型信息。 模型库的形成可以专门包括个人，“智能”模型库，差异库和预测库。 可以修改预测模型库来处理各种需求情况。

公开(公告)号：US20130230099A1

公开(公告)日：2013-09-05

申请号：US13797644

申请日：2013-03-12

申请人： EUCLID DISCOVERIES, LLC

发明人： Darin DeForest ,  Charles P. Pace ,  Nigel Lee ,  Renato Pizzorni

IPC分类号： H04N7/26

CPC分类号： H04N19/50 ,  H04N19/23 ,  H04N19/51 ,  H04N19/543 ,  H04N19/85

摘要： A model-based compression codec applies higher-level modeling to produce better predictions than can be found through conventional block-based motion estimation and compensation. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks and related to specific blocks of video data to be encoded. The tracking information is used to produce model-based predictions for those blocks of data, enabling more efficient navigation of the prediction search space than is typically achievable through conventional motion estimation methods. A hybrid framework enables modeling of data at multiple fidelities and selects the appropriate level of modeling for each portion of video data. A compliant-stream version of the model-based compression codec uses the modeling information indirectly to improve compression while producing bitstreams that can be interpreted by standard decoders.

摘要翻译： 基于模型的压缩编解码器应用较高级别的建模以产生比通过传统的基于块的运动估计和补偿可以发现的更好的预测。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道并与要编码的视频数据的特定块有关。 跟踪信息用于为这些数据块产生基于模型的预测，使得能够比通常通过常规运动估计方法可实现的预测搜索空间更有效地导航。 混合框架可以对多个保真度的数据进行建模，并为视频数据的每个部分选择适当的建模级别。 基于模型的压缩编解码器的兼容流版本间接地使用建模信息来提高压缩，同时产生可由标准解码器解释的比特流。

公开(公告)号：US20130170541A1

公开(公告)日：2013-07-04

申请号：US13772230

申请日：2013-02-20

申请人： Euclid Discoveries, LLC

发明人： Charles P. Pace ,  Darin DeForest ,  Nigel Lee ,  Renato Pizzorni ,  Richard Wingard

IPC分类号： H04N7/26

CPC分类号： H04N19/149 ,  G06T9/001

摘要： Systems and methods of improving video encoding/decoding efficiency may be provided. A feature-based processing stream is applied to video data having a series of video frames. Computer-vision-based feature and object detection algorithms identify regions of interest throughout the video datacube. The detected features and objects are modeled with a compact set of parameters, and similar feature/object instances are associated across frames. Associated features/objects are formed into tracks, and each track is given a representative, characteristic feature. Similar characteristic features are clustered and then stored in a model library, for reuse in the compression of other videos. A model-based compression framework makes use of the preserved model data by detecting features in a new video to be encoded, relating those features to specific blocks of data, and accessing similar model information from the model library. The formation of model libraries can be specialized to include personal, “smart” model libraries, differential libraries, and predictive libraries. Predictive model libraries can be modified to handle a variety of demand scenarios.

摘要翻译： 可以提供改善视频编码/解码效率的系统和方法。 基于特征的处理流被应用于具有一系列视频帧的视频数据。 基于计算机视觉的特征和对象检测算法识别整个视频数据库中的感兴趣区域。 检测到的特征和对象用一组紧凑的参数建模，并且相似的特征/对象实例在帧之间相关联。 相关特征/对象被形成轨道，并且每个轨道被给予代表性的特征。 类似的特征特征被聚类，然后存储在模型库中，以便在压缩其他视频中重用。 基于模型的压缩框架通过检测要编码的新视频中的特征来使用保留的模型数据，将这些特征与特定的数据块相关联，以及从模型库访问相似的模型信息。 模型库的形成可以专门包括个人，“智能”模型库，差异库和预测库。 可以修改预测模型库来处理各种需求情况。