-
公开(公告)号:KR101419012B1
公开(公告)日:2014-07-14
申请号:KR1020130105559
申请日:2013-09-03
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
CPC classification number: H04N19/56 , H04N19/105 , H04N19/13 , H04N19/139 , H04N19/176 , H04N19/513 , H04N19/52 , H04N19/593 , H04N19/70
Abstract: 본 발명의 영상 부호화 장치는 후보 예측 움직임 벡터 리스트에 포함된 후보 예측 움직임 벡터 중 예측 대상 블록에 대한 화면 간 예측을 수행하기 위해 사용되는 예측 움직임 벡터에 대한 정보를 부호화하는 엔트로피 부호화부, 및 상기 예측 대상 블록의 주변 예측 블록으로부터 복수개의 공간적 후보 예측 움직임 벡터 관련 정보를 결정하고 상기 복수개의 공간적 후보 예측 움직임 벡터 관련 정보를 기초로 시간적 후보 예측 움직임 벡터 관련 정보를 결정하여 상기 후보 예측 움직임 벡터 리스트를 생성하는 예측부를 포함할 수 있다. 따라서, 최적의 예측 움직임 벡터를 유도시 발생하는 복잡도를 줄이고 부호화 효율을 높일 수 있다.
Abstract translation: 公开了一种用于诱导预测运动矢量的方法及其使用方法。 图像解码方法可以包括:从预测的目标块的周边预测块确定与多个空间候选预测运动矢量相关的信息的步骤; 以及基于与多个空间候选预测运动矢量相关的信息来确定与时间候选预测运动矢量有关的信息的步骤。 因此,本发明可以降低复杂度并且可以在诱导最佳预测运动矢量时提高编码效率。
-
公开(公告)号:KR1020130111501A
公开(公告)日:2013-10-10
申请号:KR1020130105561
申请日:2013-09-03
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
CPC classification number: H04N19/56 , H04N19/105 , H04N19/13 , H04N19/139 , H04N19/176 , H04N19/513 , H04N19/52 , H04N19/593 , H04N19/70
Abstract: PURPOSE: Video encoding and decoding efficiency is increased. CONSTITUTION: A plurality of spacious candidate prediction moving vector-related information is determined from an around prediction block of a prediction object block. Time candidate prediction moving vector-related information is determined based on the plurality of spacious candidate prediction moving vector-related information. A candidate prediction moving vector list is formed based on the time candidate prediction moving vector-related information and the spacious candidate prediction moving vector-related information. A moving vector is determined based on final prediction moving vector index information among candidate prediction moving vector list. [Reference numerals] (111) Motion prediction unit; (112) Motion compensation unit; (120) Intra prediction unit; (130) Conversion unit; (140) Quantization unit; (150) Entropy encoding unit; (160) Reverse quantization unit; (170) Reverse conversion unit; (180) Filter unit; (190) Reference image buffer; (AA) Input image; (BB) Inter; (CC) Intra; (DD) Bitstream
Abstract translation: 目的:视频编码和解码效率提高。 构成:从预测对象块的周围的预测块确定多个候选预测运动矢量相关信息。 基于多个宽敞的候选预测运动矢量相关信息来确定时间候选预测运动矢量相关信息。 基于时间候选预测运动矢量相关信息和宽敞的候选预测运动矢量相关信息形成候选预测运动矢量列表。 基于候选预测运动矢量列表中的最终预测运动矢量索引信息来确定运动矢量。 (附图标记)(111)运动预测单元; (112)运动补偿单元; (120)内部预测单元; (130)转换单位; (140)量化单元; (150)熵编码单元; (160)反向量化单元; (170)反转换单元; (180)过滤单元; (190)参考图像缓冲器; (AA)输入图像; (BB)国际; (CC)内 (DD)位流
-
公开(公告)号:KR1020130004173A
公开(公告)日:2013-01-09
申请号:KR1020120071620
申请日:2012-07-02
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/176 , H04N19/119 , H04N19/50
CPC classification number: H04N19/119 , H04N19/105 , H04N19/132 , H04N19/147 , H04N19/157 , H04N19/176 , H04N19/196 , H04N19/593 , H04N19/70 , H04N19/91
Abstract: PURPOSE: A method and an apparatus for video encoding and decoding are provided to reduce complexity by using the information of neighboring LCUs(Largest Coding Units) in coding/decodeing a current LCU. CONSTITUTION: An encoder and a decoder specify the position of LCUs(Largest Coding Units) around a current LCU. The encoder and the decoder obtain the encoding information of the LCUs. Based on the encoding information obtained from the LCUs, the encoder and the decoder perform coding and decoding on the current LCU. [Reference numerals] (400) Current LCU; (AA) Bring encoding information of a surrounding LCU
Abstract translation: 目的:提供一种用于视频编码和解码的方法和装置,以便在编码/解码当前LCU时使用相邻LCU(最大编码单元)的信息来降低复杂度。 构成:编码器和解码器指定当前LCU周围的LCU(最大编码单位)的位置。 编码器和解码器获得LCU的编码信息。 基于从LCU获得的编码信息,编码器和解码器对当前的LCU执行编码和解码。 (附图标记)(400)当前LCU; (AA)带有周围LCU的编码信息
-
公开(公告)号:KR101983124B1
公开(公告)日:2019-05-29
申请号:KR1020180080808
申请日:2018-07-11
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/513 , H04N19/59 , H04N19/176
-
公开(公告)号:KR101973487B1
公开(公告)日:2019-04-29
申请号:KR1020180080801
申请日:2018-07-11
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/513 , H04N19/59 , H04N19/176
-
公开(公告)号:KR101959490B1
公开(公告)日:2019-03-19
申请号:KR1020120018672
申请日:2012-02-23
Applicant: 한국항공대학교산학협력단
IPC: H04N19/147 , H04N19/152 , H04N19/124 , H04N19/149
-
公开(公告)号:KR1020160057364A
公开(公告)日:2016-05-23
申请号:KR1020160055707
申请日:2016-05-04
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/59 , H04N19/105 , H04N19/11 , H04N19/124
CPC classification number: H04N19/157 , H04N19/11 , H04N19/176 , H04N19/198 , H04N19/593 , H04N19/70
Abstract: 인트라예측모드부호화및 복호화방법, 그리고영상복호화장치및 영상부호화장치가개시된다. 상기인트라예측모드복호화방법은현재예측단위에인접한주변예측단위로부터 MPM(Most Probable Mode)을도출하는단계및 상기도출된 MPM 중상기현재예측단위의인트라예측모드와동일한예측모드를가지는 MPM이존재하는지여부를지시하는 MPM 플래그를기반으로상기현재예측단위의인트라예측모드를도출하는단계를포함한다.
Abstract translation: 公开了一种用于帧内预测模式的编码/解码的方法,用于对图像进行解码的装置以及用于编码图像的装置。 根据本发明,该方法包括:从与当前预测单元相邻的周围预测单元中引入最可能模式(MPM)的步骤; 以及基于指示在所述诱导的MPM中是否存在具有与当前预测单元的帧内预测模式相同的预测模式的MPM的MPM标志来诱导当前预测单元的帧内预测模式的步骤。 本发明提供一种用于以增强的图像编码/解码效率对图像进行编码/解码的方法。
-
公开(公告)号:KR101620619B1
公开(公告)日:2016-05-12
申请号:KR1020130010488
申请日:2013-01-30
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/59
CPC classification number: H04N19/157 , H04N19/11 , H04N19/176 , H04N19/198 , H04N19/593 , H04N19/70
Abstract: 인트라예측모드부호화및 복호화방법, 그리고영상복호화장치및 영상부호화장치가개시된다. 상기인트라예측모드복호화방법은현재예측단위에인접한주변예측단위로부터 MPM(Most Probable Mode)을도출하는단계및 상기도출된 MPM 중상기현재예측단위의인트라예측모드와동일한예측모드를가지는 MPM이존재하는지여부를지시하는 MPM 플래그를기반으로상기현재예측단위의인트라예측모드를도출하는단계를포함한다.
-
公开(公告)号:KR1020130099852A
公开(公告)日:2013-09-06
申请号:KR1020130020447
申请日:2013-02-26
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/59
CPC classification number: H04N19/50 , H04N19/105 , H04N19/159 , H04N19/176
Abstract: PURPOSE: A method for encoding and decoding in an intra prediction mode is provided to simply induce an intra prediction mode without a complicated operation process. CONSTITUTION: If two identical induced most probable modes (MPMs) induced in adjacent prediction units are not included in at least two set prediction modes, an image encoding device sets up remaining two MPMs of an MPM list consisting of three MPMs for a current prediction unit out of the set prediction mode and a proximity prediction mode adjacent to the induced MPMs (S1001,S1003). The image encoding device encodes a prediction mode for the current prediction unit based on the induced MPMs and the set remaining two MPMs. [Reference numerals] (AA) MPM list organization; (S1001) Left MPM=Previously-defined encoding mode A or Left MPM=Previously-defined encoding mode B; (S1002) MPM[0]=Previously-defined encoding mode A, MPM[1]=Previously-defined encoding mode B, MPM[2]=Previously-defined encoding mode C; (S1003) MPM[0]=Left MPM, MPM[1]=Previously-defined encoding mode A, MPM[2]=Previously-defined encoding mode B
Abstract translation: 目的:提供一种帧内预测模式的编码和解码方法,以简单地引导帧内预测模式,而不需要复杂的操作过程。 构成:如果在至少两组预测模式中不包括在相邻预测单元中诱导的两个相同的诱导最可能模式(MPM),则图像编码装置设置由用于当前预测单元的三个MPM组成的MPM列表的剩余两个MPM (S1001,S1003)附近的邻近预测模式。 图像编码装置根据诱导的MPM和剩下的两个MPM对当前预测单元的预测模式进行编码。 (附图)(AA)MPM清单组织; (S1001)左MPM =以前定义的编码模式A或左MPM =以前定义的编码模式B; (S1002)MPM [0] =以前定义的编码模式A,MPM [1] =以前定义的编码模式B,MPM [2] =以前定义的编码模式C; (S1003)MPM [0] =左MPM,MPM [1] =以前定义的编码模式A,MPM [2] =以前定义的编码模式B
-
公开(公告)号:KR1020130030232A
公开(公告)日:2013-03-26
申请号:KR1020120102214
申请日:2012-09-14
Applicant: 한국전자통신연구원 , 한국항공대학교산학협력단
IPC: H04N19/51 , H04N19/139
CPC classification number: H04N19/56 , H04N19/105 , H04N19/13 , H04N19/139 , H04N19/176 , H04N19/513 , H04N19/52 , H04N19/593 , H04N19/70
Abstract: PURPOSE: A method for inducing a prediction MV(Motion Vector) and a device using the method are provided to reduce complexity generated when inducing the prediction MV by calculating a final prediction MV by forming a candidate prediction MV list. CONSTITUTION: An image encoding/decoding device induces temporal candidate prediction MV information based on spatial candidate prediction MV information(S1420). The device adds an induced candidate prediction MV to a candidate prediction MV list(S1430). The device removes the same candidate prediction MV to reform the candidate prediction MV list(S1440). The device determines a final prediction MV from the candidate prediction MV list(S1460). [Reference numerals] (AA) Start; (BB) End; (S1400) Guiding spatial candidate prediction motion vectors; (S1410) Determining whether two induced spatial candidate prediction motion vectors are different from each other; (S1420) Inducing temporal candidate prediction motion vectors; (S1430) Adding the induced candidate prediction motion vectors to a candidate prediction motion vector list; (S1440) Reconstructing a candidate prediction motion vector list by removing identical candidate prediction motion vectors; (S1450) Adjusting the size of the candidate prediction motion vector list by adding and removing candidate prediction motion vectors; (S1460) Determining the final prediction motion vector from the candidate prediction motion vector list
Abstract translation: 目的:提供一种用于诱导预测MV(运动矢量)的方法和使用该方法的装置,以通过形成候选预测MV列表来计算最终预测MV,来降低当诱导预测MV时产生的复杂度。 构成:图像编码/解码装置基于空间候选预测MV信息诱导时间候选预测MV信息(S1420)。 该装置将候选预测MV列表添加到候选预测MV列表中(S1430)。 该装置移除相同的候选预测MV,以改变候选预测MV列表(S1440)。 该装置从候选预测MV列表确定最终预测MV(S1460)。 (附图标记)(AA)开始; (BB)结束; (S1400)引导空间候选预测运动矢量; (S1410)确定两个诱导空间候选预测运动矢量是否彼此不同; (S1420)诱导时间候选预测运动矢量; (S1430)将候选预测运动矢量添加到候选预测运动矢量列表中; (S1440)通过去除相同的候选预测运动矢量来重构候选预测运动矢量列表; (S1450)通过添加和删除候选预测运动矢量来调整候选预测运动矢量列表的大小; (S1460)从候选预测运动矢量列表确定最终预测运动矢量
-
-
-
-
-
-
-
-
-