公开(公告)号：US06834080B1

公开(公告)日：2004-12-21

申请号：US09655452

申请日：2000-09-05

申请人： Rieko Furukawa ,  Takeshi Nagai ,  Yoshihiro Kikuchi

发明人： Rieko Furukawa ,  Takeshi Nagai ,  Yoshihiro Kikuchi

IPC分类号： H04N712

CPC分类号： H04N19/198 ,  H04N19/102 ,  H04N19/124 ,  H04N19/132 ,  H04N19/139 ,  H04N19/14 ,  H04N19/15 ,  H04N19/17 ,  H04N19/172 ,  H04N19/176 ,  H04N19/179 ,  H04N19/192 ,  H04N19/196 ,  H04N19/527 ,  H04N19/587 ,  H04N19/61 ,  H04N19/87

摘要： An video encoding method and a video encoding apparatus wherein an input video signal is divided into a plurality of temporally continuous scenes each including at least one frame, statistical feature amounts are calculated for each of the scenes, encoded parameters are generated for each of the scenes based on said statistical feature amounts, and the input image signal is encoded using the encoded parameters.

摘要翻译： 一种视频编码方法和视频编码装置，其中输入视频信号被分成多个时间上连续的场景，每个场景包括至少一个帧，对于每个场景计算统计特征量，为每个场景生成编码参数 基于所述统计特征量，并且使用编码参数对输入图像信号进行编码。

公开(公告)号：US06829301B1

公开(公告)日：2004-12-07

申请号：US09092225

申请日：1998-06-05

申请人： Michael Tinker ,  Jeremy D. Pollack ,  Glenn Arthur Reitmeier

发明人： Michael Tinker ,  Jeremy D. Pollack ,  Glenn Arthur Reitmeier

IPC分类号： H04N712

CPC分类号： H04N19/59 ,  H04N19/103 ,  H04N19/186 ,  H04N19/187 ,  H04N19/20 ,  H04N19/30 ,  H04N19/61 ,  H04N19/63 ,  H04N19/98

摘要： A method and concomitant apparatus for compressing, multiplexing and, in optional embodiments, encrypting, transporting, decrypting, decompressing and presenting high quality video information in a manner that substantially preserves the fidelity of the video information in a system utilizing standard quality circuits to implement high quality compression, transport and decompression.

摘要翻译： 一种用于压缩，多路复用的方法和附带装置，并且在可选实施例中，以基本上保持视频信息在系统中的保真度的方式来加密，传输，解密，解压缩和呈现高质量视频信息，该系统利用标准质量电路来实现高 质量压缩，运输和减压。

公开(公告)号：US06829300B1

公开(公告)日：2004-12-07

申请号：US09786590

申请日：2001-03-05

申请人： Hidetoshi Ichioka

发明人： Hidetoshi Ichioka

IPC分类号： H04N712

CPC分类号： H03M7/40 ,  H03M7/46 ,  H04N19/60 ,  H04N19/91

摘要： A recompression program 23 being a video edition software 20 executed by a personal computer 17 on a compressed video data supplied from a DV camera deck 11 or hard disc drive 12 and variable-length coded for conformance to the DV specification, compresses a variable-length code word by replacing one code word of a code word string of the variable-length code words with some other code word shorter than the one cord word and having a value approximate to that of the one code word.

摘要翻译： 作为由个人计算机17对从DV摄像机机架11或硬盘驱动器12提供的压缩视频数据执行的视频编辑软件20的再压缩程序23，其可变长度被编码以符合DV规范，压缩可变长度 通过将可变长度代码字的代码字串的一个代码字替换为比一个索引词短的一些其他代码字并具有与该代码字的值近似的值来代码字。

公开(公告)号：US06826231B2

公开(公告)日：2004-11-30

申请号：US09973749

申请日：2001-10-11

申请人： Kuniaki Takahashi ,  Kazushi Sato ,  Teruhiko Suzuki

发明人： Kuniaki Takahashi ,  Kazushi Sato ,  Teruhiko Suzuki

IPC分类号： H04N712

CPC分类号： H04N19/577 ,  H04N19/105 ,  H04N19/107 ,  H04N19/112 ,  H04N19/132 ,  H04N19/137 ,  H04N19/176 ,  H04N19/40 ,  H04N19/46 ,  H04N19/51 ,  H04N19/523 ,  H04N19/56 ,  H04N19/587 ,  H04N19/59 ,  H04N19/61

摘要： The invention provides a motion vector conversion method by which the coding efficiency in image coding of MPEG4 in an image information conversion method can be further augmented. In the motion vector conversion method for an image information conversion method, every other one of P frames of a bit stream of MPEG2 is dropped to produce a bit stream of MPEG4 of a reduced frame rate and a low bit rate. Also for motion vector modification, information of each of macro blocks and between blocks in the inputted bit stream representative of image compression information of MPEG2 is stored in advance, and motion vectors of a P frame immediately preceding to each P frame to be dropped are duplicated based on the stored information. Then, the duplicated motion vectors are extended to twice in the temporal direction to produce motion vectors of MPEG4 converted from an intraframe of MPEG2.

摘要翻译： 本发明提供一种运动矢量转换方法，通过该方法可以进一步增强图像信息转换方法中MPEG4的图像编码的编码效率。 在用于图像信息转换方法的运动矢量转换方法中，丢弃MPEG2比特流的P帧中的每一个以产生具有降低的帧速率和低比特率的MPEG4的比特流。 此外，对于运动矢量修改，预先存储代表MPEG2的图像压缩信息的输入比特流中的每个宏块和块之间的信息，并且复制每个待丢弃的P帧之前的P帧的运动矢量 基于存储的信息。 然后，复制的运动矢量在时间方向上扩展到两次，以产生从MPEG2的帧内转换的MPEG4的运动矢量。

公开(公告)号：US06826230B1

公开(公告)日：2004-11-30

申请号：US08688917

申请日：1996-07-31

申请人： Philippe Salembier

发明人： Philippe Salembier

IPC分类号： H04N712

CPC分类号： G06T9/20 ,  H04N19/17 ,  H04N19/20 ,  H04N19/46 ,  H04N19/537 ,  H04N19/543

摘要： The invention relates to an image sequence coding method comprising a segmentation of each image into partitions made of regions, an estimation and compensation of the motion of these regions between two partitions, and a region contour and texture coding step. The proposed motion estimation step of this method is carried out in a loop circuit (91, 92) in which, after a definition of the mode of transmission of the regions, a particular coded signal is generated for the background regions of the partition, said signal including an order information, indicating a priority between contiguous regions, and an associated updated motion information. The implementation of this method reduces drastically the computational load before the contour coding. The invention also relates to the corresponding decoding method and system.

摘要翻译： 本发明涉及一种图像序列编码方法，包括将每个图像分割成由区域构成的分区，估计和补偿两个分区之间的这些区域的运动，以及区域轮廓和纹理编码步骤。 该方法的所提出的运动估计步骤在环路电路（91,92）中进行，其中在对区域的传输模式进行定义之后，为分区的背景区域生成特定的编码信号，所述 信号，其包括指示连续区域之间的优先级的顺序信息和相关联的更新的运动信息。 该方法的实现在轮廓编码之前大大降低了计算量。 本发明还涉及相应的解码方法和系统。

公开(公告)号：US06823011B2

公开(公告)日：2004-11-23

申请号：US09997497

申请日：2001-11-19

申请人： Ajay Divakaran ,  Regunathan Radhakrishnan

发明人： Ajay Divakaran ,  Regunathan Radhakrishnan

IPC分类号： H04N712

CPC分类号： H04N19/142 ,  G06K9/00261 ,  G06K9/00335 ,  G06K9/00744 ,  H04N19/179 ,  H04N19/46 ,  H04N19/503 ,  H04N19/517 ,  H04N19/87

摘要： A method detects an unusual event in a video. Motion vectors are extracted from each frame in a video acquired by a camera of a scene. Zero run-length parameters are determined for each frame from the motion vectors. The zero run-length parameters are summed over predetermined time intervals of the video, and a distance is determined between the sum of the zero run-lengths of a current time interval and the sum of the zero run-lengths of a previous time interval. Then, the unusual event is detected if the distance is greater than a predetermined threshold.

摘要翻译： 一种方法检测视频中的异常事件。 在由场景的相机获取的视频中的每个帧中提取运动矢量。 从运动矢量为每个帧确定零游程长度参数。 零游程长度参数在视频的预定时间间隔上相加，并且在当前时间间隔的零游程长度与前一时间间隔的零游程长度之和之间确定距离。 然后，如果距离大于预定阈值，则检测到异常事件。

公开(公告)号：US06819715B2

公开(公告)日：2004-11-16

申请号：US09910906

申请日：2001-07-24

申请人： Takahiro Nishi ,  Shintarou Nakatani

发明人： Takahiro Nishi ,  Shintarou Nakatani

IPC分类号： H04N712

CPC分类号： H04N19/61 ,  H04N19/895

摘要： When coded image data which are obtained by coding image data by a coding method including an inter-frame predictive coding process are decoded, even when the decoding is started from an inter-frame predictive coded frame, the display of a decoded image having few disturbances is started, without the display being considerably delayed from the timing of input of coded image data. An image decoding apparatus 100a for decoding coded image data and displaying a decoded image on the basis of decoded image data comprises a frame memory 112a which holds decoded image data 107 of an already processed frame as reference data which are to be used for the decoding process for the inter-frame predictive coded frame, and initializes the storage area of the frame memory 112a at the initiation of the decoding process for the coded image data.

摘要翻译： 当通过包括帧间预测编码处理的编码方法对图像数据进行编码而获得的编码图像数据被解码时，即使当从帧间预测编码帧开始解码时，显示具有较小干扰的解码图像 开始，而不会从输入编码图像数据的定时显着延迟显示。 用于解码编码图像数据并基于解码图像数据显示解码图像的图像解码装置100a包括：帧存储器112a，其将已处理帧的解码图像数据107保存为用于解码处理的参考数据 对于帧间预测编码帧，并且在编码图像数据的解码处理开始时初始化帧存储器112a的存储区域。

公开(公告)号：US06819712B2

公开(公告)日：2004-11-16

申请号：US09959458

申请日：2001-10-26

申请人： Yuji Kawashima ,  Atsushi Asano

发明人： Yuji Kawashima ,  Atsushi Asano

IPC分类号： H04N712

CPC分类号： H04N21/44004 ,  H04N19/60 ,  H04N19/61 ,  H04N19/70 ,  H04N19/91 ,  H04N21/23406 ,  H04N21/234354 ,  H04N21/23611

摘要： A picture coding apparatus has a signal processor 10 for generating macro-block information by calculating and quantizing the DCTs of a picture signal of a given unit are calculated for each macro-block MB. This information is input to a syntax generator 12 to generate syntax, which is coded by a variable-length coding processor 13. Quantized DCT coefficients of macro information before coding are counted by a count unit 21, and a code size after coding the macro-block information is counted by a count unit 14. A calculation unit 15 calculates an average code length from the accumulated value of the two code sizes. An estimation unit 14 estimates a generation code size after coding given macro-block information input to the syntax generator 12 on the basis of the average code size. A calculation unit 17 calculates a stuffing size by comparing the estimated generation code size with an actually generated generation code size, and the syntax generator 12 generates syntax on the basis of this stuffing size and macro-block MB information.

摘要翻译： 图像编码装置具有通过对每个宏块MB计算给定单位的图像信号的DCT计算和量化来产生宏块信息的信号处理器10。 该信息被输入到语法生成器12，以生成由可变长度编码处理器13编码的语法。编码之前的宏信息的量化DCT系数由计数单元21进行计数， 块信息由计数单元14计数。计算单元15根据两个码字大小的累计值计算平均码长度。 估计单元14基于平均代码大小来估计输入到语法生成器12的给定宏块信息的编码之后的生成代码大小。 计算单元17通过将估计的生成代码大小与实际生成的代码大小进行比较来计算填充大小，并且语法生成器12基于该填充大小和宏块MB信息来生成语法。

公开(公告)号：US06816553B1

公开(公告)日：2004-11-09

申请号：US09040561

申请日：1998-03-18

申请人： Hirohisa Yamaguchi

发明人： Hirohisa Yamaguchi

IPC分类号： H04N712

CPC分类号： H04N19/152 ,  H04N19/122 ,  H04N19/176 ,  H04N19/46 ,  H04N19/61

摘要： To provide high quality coding with respect to both interlace video and progressive video. It is a coding method for image signals, which includes the steps of dividing the picture elements within 1 macroblock of input image signals into multiple groups, totalling each group as interlace video by executing DCT (block 31), comparing said totalled result and the result of DCT (block 32) executed with respect to the entire picture elements within said 1 macroblock, and selecting one of the DCT results (block 34).

公开(公告)号：US06813313B2

公开(公告)日：2004-11-02

申请号：US09839924

申请日：2001-04-20

申请人： Peng Xu ,  Shih-Fu Chang ,  Ajay Divakaran

发明人： Peng Xu ,  Shih-Fu Chang ,  Ajay Divakaran

IPC分类号： H04N712

CPC分类号： G06K9/00711

摘要： A system and method analyzes a compressed video including a sequence of frames. The amount of a dominant feature in each frame of the compressed video is measured. A label is associated with each frame according the measured amount of the dominant feature. Views in the video are identified according to the labels, and the video is segmented into actions according to the views. The video can then be analyzed according to the action to determine significant events in the video.

摘要翻译： 系统和方法分析包括帧序列的压缩视频。 测量压缩视频的每个帧中的主要特征的量。 根据测量的主要特征量，标签与每个帧相关联。 根据标签识别视频中的视图，并根据视图将视频分割为动作。 然后可以根据动作分析视频以确定视频中的重要事件。