公开(公告)号：KR101538704B1

公开(公告)日：2015-07-28

申请号：KR1020090006723

申请日：2009-01-28

Applicant: 삼성전자주식회사 ,  광운대학교 산학협력단

Inventor： 조대성 ,  최웅일 ,  송관웅 ,  주영훈 ,  김용석 ,  심동규 ,  남정학 ,  박시내 ,  정광수

IPC: H04N19/80

CPC classification number: H04N19/124 ,  H04N19/117 ,  H04N19/126 ,  H04N19/50 ,  H04N19/523 ,  H04N19/61

Abstract: 이러한본 발명은, 입력영상블록의문맥정보를고려하여예측블록을생성하기위한보간필터를적응적으로선택하는선택기와, 상기문맥정보에대응하여다수의보간필터들을그룹별로저장하고있는필터뱅크와, 상기선택기의제어에따라상기필터뱅크로부터선택되어, 상기입력영상블록과관련된참조영상블록을보간하는특정보간필터를포함하는예측블록생성부를구비함을특징으로한다.

公开(公告)号：KR101517100B1

公开(公告)日：2015-05-15

申请号：KR1020080082014

申请日：2008-08-21

Applicant: 삼성전자주식회사 ,  광운대학교 산학협력단

Inventor： 송관웅 ,  이창현 ,  주영훈 ,  김용석 ,  심동규 ,  남정학

IPC: H04N19/186 ,  H04N19/51

Abstract: 영상을 부호화하는 방법에 있어서, 부호화하고자 하는 영상을 취득하는 과정과, 취득한 영상을 각 색 성분별로 분리하는 과정과, 분리된 각 색 성분별로 예측 영상을 생성하여, 예측 영상과 상기 영상 간의 차분 영상을 생성하는 과정과, 차분 영상을 미리 설정된 변환식에 적용하여 각 색 성분별로 변환 부호화를 수행하는 과정을 포함함을 특징으로 한다.
부호화, 복호화, 변환 계수

公开(公告)号：KR1020140048807A

公开(公告)日：2014-04-24

申请号：KR1020130119609

申请日：2013-10-08

Applicant: 광운대학교 산학협력단

Inventor： 유은경 ,  남정학 ,  조현호 ,  심동규

IPC: H04N19/50 ,  H04N19/00

CPC classification number: H04N19/13

Abstract: Disclosed are a method for initiating entropy for encoding/decoding an extension video and an apparatus thereof. The method for initiating entropy for decoding an extension video includes the steps of: determining whether to perform entropy initialization based on an interlayer prediction; and initializing entropy for an enhancement layer referring to information about an entropy decoding of the reference layer when it is determined that the entropy initialization is performed based on an interlayer prediction. Thus, a parameter for initializing the entropy of the reference layer is allowed to be utilized in the enhancement layer so that the coding and decoding efficiency of the entropy can be improved. [Reference numerals] (AA) Start; (BB) NO; (CC) YES; (DD) End; (S310) Entropy is initialized based on an interlayer prediction?; (S320) Extract a parameter for initializing interlayer entropy; (S330) Initialize entropy for an improved layer with reference to a parameter for initializing the interlayer entropy; (S340) Initialize general entropy for the enhancement layer; (S350) Decode the entropy for the enhancement layer

Abstract translation: 公开了一种用于启动用于对扩展视频进行编码/解码的熵的方法及其装置。 用于启动用于解码扩展视频的熵的方法包括以下步骤：基于层间预测确定是否执行熵初始化; 以及当确定基于层间预测执行熵初始化时，参考关于参考层的熵解码的信息来初始化增强层的熵。 因此，允许在增强层中利用用于初始化参考层的熵的参数，从而可以提高熵的编码和解码效率。 （附图标记）（AA）开始; （BB）NO; （CC）是; （DD）结束; （S310）基于层间预测δ来初始化熵; （S320）提取初始化夹层熵的参数; （S330）参考用于初始化层间熵的参数来初始化改进层的熵; （S340）初始化增强层的一般熵; （S350）对增强层的熵进行解码

公开(公告)号：KR1020140048806A

公开(公告)日：2014-04-24

申请号：KR1020130119606

申请日：2013-10-08

Applicant: 광운대학교 산학협력단

Inventor： 조현호 ,  유은경 ,  남정학 ,  심동규

IPC: H04N19/50

CPC classification number: H04N19/50 ,  H04N19/105 ,  H04N19/132

Abstract: Disclosed are an interlayer prediction method for a multilayer video based on a spatial resolution and an apparatus thereof. A decoding apparatus using an interlayer prediction based on a spatial resolution includes an adding unit for adding a prediction value to a residual value of a reference layer to generate a restored picture of the reference layer; a reference point determining unit for determining a reference point representing a step of processing the restored picture of the reference layer which a current picture of an enhancement layer refers to; and an interpolating unit for generating an interpolation picture of the reference layer by interpolating the restored picture of the reference layer at the reference point based on a resolution of the current picture of the enhancement layer. Thus, pixel values at various points are used as input values of the interpolating unit in the reference layer, so that the interpolation can be more effectively performed according to image and coding conditions. [Reference numerals] (110,210) Entropy decoding unit; (120,220) Inverse quantization unit; (130,230) Inversion unit; (150,250) Deblocking filter unit; (160,260) SAO performing unit; (170,270) Frame memory; (180,280) Intra prediction unit; (190,290) Inter prediction unit; (300) Reference point determining unit; (400) Interpolation unit; (500) Interlayer SAO performing unit; (AA,BB) Bit stream

Abstract translation: 公开了一种基于空间分辨率的多层视频的层间预测方法及其装置。 使用基于空间分辨率的层间预测的解码装置包括用于将预测值与参考层的残差值相加以生成参考层的恢复图像的相加单元; 参考点确定单元，用于确定表示加强层的当前图像所参考的参考层的恢复图像的步骤的参考点; 以及内插单元，用于通过基于增强层的当前图像的分辨率内插参考点上的参考层的恢复图像来生成参考层的内插图像。 因此，将各点的像素值作为参照层内插单位的输入值，能够根据图像和编码条件更有效地进行插值。 （标号）（110,210）熵解码单元; （120,220）反量化单元; （130,230）反转单位; （150,250）去块过滤器单元; （160,260）SAO执行单位; （170,270）帧记忆; （180,280）内部预测单位; （190,290）预测单位; （300）参考点确定单元; （400）插值单元; （500）中间层SAO执行单元; （AA，BB）位流

公开(公告)号：KR1020130070191A

公开(公告)日：2013-06-27

申请号：KR1020110137398

申请日：2011-12-19

Applicant: 광운대학교 산학협력단

Inventor： 조현호 ,  유은경 ,  남정학 ,  심동규

IPC: H04N19/82 ,  H04N19/86

CPC classification number: H04N19/82 ,  H04N19/436 ,  H04N19/86

Abstract: PURPOSE: A parallel in-loop filtering method of an LCU(Largest Coding Unit) and a device thereof are provided to include a pixel storage unit of the LCU and a storage pixel load unit of the LCU unit, thereby reducing dependability with surrounding LCUs when a deblocking filter, SAO(Sample Adaptive Offset), and an ALF(Adaptive Loop Filter) are performed by using the LCU. CONSTITUTION: A pixel storage unit(410) of an LCU stores pixels of the LCU in a buffer before filtering when a deblocking filter, SAO, and an ALF are performed by using the LCU. A storage pixel load unit(415) of the LCU unit loads pixel values of surrounding LCUs, which are stored in the buffer, to a padding area before a parallel in-loop filtering process of the LCU unit. [Reference numerals] (410) Pixel storage unit of an LCU unit; (415) Stored pixel load unit of the LCU unit; (420) Deblocking filter performing unit for a vertical boundary; (425) Deblocking filter performing unit for a horizontal boundary; (430) SAO performing unit of the LCU unit; (435) ALF performing unit of the LCU unit

Abstract translation: 目的：提供LCU（最大编码单元）的并行环路滤波方法及其装置，以包括LCU的像素存储单元和LCU单元的存储像素负载单元，由此降低与周围LCU的可靠性， 通过使用LCU来执行解块滤波器SAO（采样自适应偏移）和ALF（自适应环路滤波器）。 构成：当通过使用LCU执行去块滤波器SAO和ALF时，LCU的像素存储单元（410）在滤波之前将LCU的像素存储在缓冲器中。 LCU单元的存储像素负载单元（415）将存储在缓冲器中的周围LCU的像素值加载到LCU单元的并行循环滤波处理之前的填充区域。 （410）LCU单元的像素存储单元; （415）LCU单元的存储像素负载单元; （420）用于垂直边界的去块滤波器执行单元; （425）水平边界的去块滤波器执行单元; （430）LCU单元的SAO执行单元; （435）LCU单元的ALF执行单元

公开(公告)号：KR1020140072941A

公开(公告)日：2014-06-16

申请号：KR1020120139411

申请日：2012-12-04

Applicant: 광운대학교 산학협력단

Inventor： 유은경 ,  남정학 ,  심동규

IPC: H04N19/59

CPC classification number: H04N19/59

Abstract: The present invention relates to image processing technology and, more specifically, a method and a device for inter-prediction in encoding and decoding of an extended video. The present invention includes a step of generating a prediction image through a restored pixel in an enhancement layer and a sample in a reference layer by loading the sample of the reference layer when performing an intra-prediction of the enhancement layer by an encoder and a decoder for the extended video.

Abstract translation: 本发明涉及图像处理技术，更具体地，涉及一种用于在扩展视频的编码和解码中进行帧间预测的方法和装置。 本发明包括以下步骤：当通过编码器和解码器执行增强层的帧内预测时，通过加载参考层的样本，通过增强层中的恢复像素和参考层中的采样来生成预测图像 用于扩展视频。

公开(公告)号：KR1020140072937A

公开(公告)日：2014-06-16

申请号：KR1020120139391

申请日：2012-12-04

Applicant: 광운대학교 산학협력단

Inventor： 유은경 ,  남정학 ,  심동규

IPC: H04N19/50

CPC classification number: H04N19/50 ,  H04N19/176

Abstract: The present invention relates to a device and a method for improving encoding efficiency by improving a block boundary after estimating the block boundary in a general video encoding/decoding device and an expansion video encoding/decoding device. According to an embodiment of the present invention, the method for improving a block boundary of a video encoding/decoding device includes a step of loading an estimated image and/or a block; a step of setting the block boundary and/or the estimated image; and a step of correcting the estimated image and/or the boundary.

Abstract translation: 本发明涉及一种在通用视频编码/解码装置和扩展视频编码/解码装置中估计块边界之后通过改进块边界来提高编码效率的装置和方法。 根据本发明的实施例，用于改进视频编码/解码装置的块边界的方法包括加载估计图像和/或块的步骤; 设置块边界和/或估计图像的步骤; 以及校正估计图像和/或边界的步骤。

公开(公告)号：KR1020140048804A

公开(公告)日：2014-04-24

申请号：KR1020130119601

申请日：2013-10-08

Applicant: 광운대학교 산학협력단

Inventor： 남정학 ,  유은경 ,  조현호 ,  심동규

IPC: H04N19/00 ,  H04N13/00

CPC classification number: H04N19/597 ,  H04N13/271

Abstract: Disclosed are an apparatus and a method for controlling a bit rate for encoding/decoding a three-dimensional image. The apparatus for encoding a multi-view image includes an extending time point bit rate control unit for determining quantization parameter for a current block of at an extending time point by calculating complexity of a current block at the extending time point referring to complexity of a reference block at a base time point corresponding to the current block at an extending time point, and a quantizing unit for performing quantization by using a determined quantization parameter. Thus, complexity can be exactly predicted so that the bit rate control can be more exactly performed. [Reference numerals] (100) Base time point bit rate control unit; (100-1) Extending time point bit rate control unit; (120,120-1) Converting unit; (121,121-1) Inversion unit; (130,130-1) Quantization unit; (131,131-1) Inverse quantization unit; (140,140-1) Entropy encoding unit; (160,160-1) In-loop filter; (170,170-1) Frame memory; (180,180-1) Intra prediction unit; (190,190-1) Motion compensation unit; (300) Multiplexer; (AA) Extending time point image; (BB) Bit stream; (CC) Base time point image

Abstract translation: 公开了一种用于控制用于对三维图像进行编码/解码的比特率的装置和方法。 用于编码多视图图像的装置包括扩展时间点比特率控制单元，用于通过根据参考的复杂度计算延长时间点的当前块的复杂度来确定在延长时间点的当前块的量化参数 在延长时间点对应于当前块的基本时间点处的块;以及量化单元，用于通过使用确定的量化参数来执行量化。 因此，可以精确地预测复杂度，使得可以更精确地执行比特率控制。 （附图标记）（100）基本时间点比特率控制单元; （100-1）扩展时间点比特率控制单元; （120,120-1）转换单元; （121,121-1）反转单位; （130,130-1）量化单元; （131,131-1）反量化单元; （140,140-1）熵编码单元; （160,160-1）环路滤波器; （170,170-1）帧存储器; （180,180-1）内部预测单位; （190,190-1）运动补偿单元; （300）多路复用器 （AA）延长时间点图像; （BB）比特流; （CC）基准时间点图像

公开(公告)号：KR1020130087980A

公开(公告)日：2013-08-07

申请号：KR1020120011969

申请日：2012-02-06

Applicant: 광운대학교 산학협력단

Inventor： 최효민 ,  심동규 ,  남정학

IPC: H04N19/50

CPC classification number: H04N19/597 ,  H04N19/124 ,  H04N19/13 ,  H04N19/159 ,  H04N19/513 ,  H04N19/80

Abstract: PURPOSE: A scalable multi-view video decoding device and a method thereof are provided to adaptively perform encoding and decoding processes according to a single-loop decoding method and a multiple-loop decoding method, and supply scalable and multi-view encoding together, thereby offering more various and expanded encoding functions. CONSTITUTION: A first decoding unit (115) generates a first differential image by decoding the input image of a first channel. A first prediction unit (130) generates a first prediction image with reference to a reference image. If the input image of the first channel corresponds to a multiple loop mode, or if the input image corresponds to a single loop mode and the header information corresponding to the input image shows an intra type, an interpolated image generating unit (135) interpolates a first restored image generated by adding the first differential image and a second prediction image. [Reference numerals] (110) Receiving unit; (115) First decoding unit; (120) First filter; (125) First buffer; (130) First prediction unit; (135) First interpolated image generating unit; (140) Second decoding unit; (145) Second filter; (150) Second buffer; (155) Second prediction unit; (160) Second interpolated image generating unit; (165) Third decoding unit; (170) Third filter; (175) Third buffer; (180) Third prediction unit

Abstract translation: 目的：提供可扩展的多视点视频解码装置及其方法，以根据单循环解码方法和多循环解码方法自适应地执行编码和解码处理，并且一起提供可缩放和多视图编码，由此 提供更多的各种和扩展的编码功能。 构成：第一解码单元（115）通过对第一信道的输入图像进行解码来生成第一差分图像。 第一预测单元（130）参考参考图像生成第一预测图像。 如果第一通道的输入图像对应于多循环模式，或者如果输入图像对应于单​​循环模式，并且对应于输入图像的标题信息表示帧内类型，则内插图像生成单元（135）内插 通过添加第一差分图像和第二预测图像而生成的第一恢复图像。 （附图标记）（110）接收单元; （115）第一解码单元; （120）第一过滤器; （125）第一缓冲液; （130）第一预测单元; （135）第一内插图像生成单元; （140）第二解码单元; （145）第二过滤器; （150）第二缓冲器; （155）第二预测单元; （160）第二内插图像生成单元; （165）第三解码单元; （170）第三过滤器; （175）第三缓冲液; （180）第三预测单元

公开(公告)号：KR1020130078569A

公开(公告)日：2013-07-10

申请号：KR1020110147588

申请日：2011-12-30

Applicant: 광운대학교 산학협력단

Inventor： 임웅 ,  남정학 ,  유종훈 ,  오승준 ,  심동규

IPC: H04N19/167 ,  H04N19/59 ,  H04N19/105 ,  G06T9/00 ,  H04N19/82 ,  H04N21/234

CPC classification number: H04N19/167 ,  G06T9/004 ,  H04N19/105 ,  H04N19/59 ,  H04N19/82 ,  H04N21/234

Abstract: PURPOSE: A method and a device of coding and decoding screen content video for improving region of interest (ROI) based screen quality are provided to improve subjective quality of the screen content video by determining an ROI considering screen content properties, assigning more bits in a main ROI, and reducing information amount assigned in another regions. CONSTITUTION: An ROI extraction unit (201) extracts ROI information on reference to an inputted original image. An image encoder (202) applies encoding methods according to an interest rate of each region on reference to the extracted ROI information. A rate-distortion adjustment unit (203) adjusts the quality of whole images and a generated bit rate and adjusts subjective quality and the generated bit rate according to the interest rate. The ROI information of a corresponding region is added to an output bit stream (204) when the coding is performed. [Reference numerals] (200) Input image; (201) ROI extraction unit; (202) Image encoder; (203) Rate-distortion adjustment unit; (204) Output bit stream

Abstract translation: 目的：提供一种编码和解码屏幕内容视频的方法和装置，用于改善基于感兴趣区域（ROI）的屏幕质量，以通过根据屏幕内容属性确定ROI来提高屏幕内容视频的主观质量，分配更多位 主要投资回报率，并减少在其他地区分配的信息量。 构成：ROI提取单元（201）参考输入的原始图像提取ROI信息。 图像编码器（202）参考所提取的ROI信息，根据每个区域的利率来应用编码方法。 速率失真调整单元（203）调整整个图像的质量和生成的比特率，并且根据利率来调整主观质量和生成的比特率。 当执行编码时，相应区域的ROI信息被添加到输出比特流（204）。 （附图标记）（200）输入图像; （201）ROI提取单元; （202）图像编码器; （203）速率失真调整单元; （204）输出位流