WO2006108780A1 - Verfahren und vorrichtung zur reduktion eines quantisierungsfehlers - Google Patents
Verfahren und vorrichtung zur reduktion eines quantisierungsfehlers Download PDFInfo
- Publication number
- WO2006108780A1 WO2006108780A1 PCT/EP2006/061312 EP2006061312W WO2006108780A1 WO 2006108780 A1 WO2006108780 A1 WO 2006108780A1 EP 2006061312 W EP2006061312 W EP 2006061312W WO 2006108780 A1 WO2006108780 A1 WO 2006108780A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- correction value
- coefficient
- quantized
- quantization
- characteristic
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/136—Incoming video signal characteristics or properties
- H04N19/14—Coding unit complexity, e.g. amount of activity or edge presence estimation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
- H04N19/124—Quantisation
- H04N19/126—Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
- H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
- H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
Definitions
- the invention relates to a method according to the preamble of claim 1 and a device according to the preamble of claim 18.
- lossless compression techniques such as Lempel-Ziv coding or Huffman coding
- the original, non-compressed information can be reconstructed without error from compressed information.
- lossy compression techniques such as the JPEG (JPEG Joint Picture Expert Group) method
- MPEG2 method MPEG-Motion Picture Expert Group
- AAC method AAC-Adaptive Audio Coding
- FIG. 1 exemplarily shows a simplified encoding unit of a JPEG encoder JC.
- an image IM to be compressed is divided into image blocks BB having, for example, 8 ⁇ 8 pixels. Further processing within the JPEG encoder JC is based on these image blocks.
- Each image block is first transformed into the frequency domain by means of the discrete cosine transform FDCT.
- each coefficient X is subjected to a quantization FQ, wherein the quantization, i. the quantization factor Q is controlled by means of the first table TS1.
- the quantized coefficient Z is generated.
- entropy coding e.g. a Huffman coding
- EC entropy encoder
- FIG. 2 shows by way of example a simplified representation of a decoding unit of a JPEG decoder JD.
- the coded image data is read from the file JDS and supplied to the entropy decoder ED for entropy decoding.
- the entropy decoder ED is controlled by the second table TS2.
- the quantized coefficients Z are available at the output of the entropy decoder. These are then inversely quantized by means of an inverse quantization module FIQ, wherein the first table TS1 controls the inverse quantization.
- the inverse quantization module FIQ supplies reconstructed coefficients Y. Finally, these are transformed by means of the inverse discrete cosine transformation IDCT from the frequency domain into the location domain and stored in the reconstructed image IM 1 at the corresponding spatial position.
- Quantization is an often used method within lossy compression techniques. Using the following equation, the function of the quantization can be represented:
- the coefficients X are represented by 8 bits, then the coefficient X can assume a value in the number range from 0 to 255.
- a size of the number space of the quantized coefficients Z is reduced as a function of the quantization factor Q.
- the quantized coefficients Z can only take a numerical value from 0 to 15. Larger quantization factors Q increase the compression rate.
- this equation corresponds to an inverse quantization and the reference symbol Y represents the reconstructed value or reconstructed coefficients.
- a first quantization error QF1 results which can be numerically determined, for example, by means of the following equation:
- Equation (3) represents only one possible type of calculation for the first quantization error QF1.
- Equation (3) represents only one possible type of calculation for the first quantization error QF1.
- a correction value can be set within the inverse quantization. 200506343
- equation (4) provides a low quantization error only for equally distributed coefficients X within the quantization interval determined by the quantization factor Q. For not uniformly distributed coefficients X, equation (4) does not provide a minimum quantization error.
- the invention has for its object to provide a method and an apparatus that reduces a quantization error in a compression method over the prior art in a simple manner.
- the quantization error describes a difference between the quantized coefficient after the inverse quantization and a corresponding coefficient before quantization
- at least one for the compression of the quanti ⁇ overbased coefficient characteristic parameter is selected, wherein at least two characteristic parameters from ⁇ be selected if one of the two characteristic parameters corresponding to a temporal prediction mode, in dependence of at least one selected charac ⁇ rule parameter, a correction value is selected, the inverse quantization after addition of the correction value and the quantized coefficient.
- an optimal correction value can be found which takes into account statistical properties of the at least one characteristic parameter with respect to the quantized coefficients and thus yields a smaller quantization error than without this consideration. Furthermore, by considering a plurality of characteristic parameters, a respective optimum correction value can be found which gives a very small quantization error for a quantized coefficient which is coded with these characteristic parameters. In general, the quantization error decreases as the characteristic parameters considered increase. Further, the determination may be the optimal or the correction values are performed offline, so that during the transit performing the compression process, a lower proces ⁇ maintenance effort for selecting the correction value is needed. Offline in this context means that in particular with the help of one or more test sequences for a 200506343
- optimal correction ⁇ values are determined. These determined optimal correction values can then be used in the inverse quantization of the quantized coefficient for the rapid determination of the correction value.
- the low Swisssauf ⁇ wall is especially important for small devices such as bilfunktelefonen at Mo ⁇ , because it attracts a great advantage low power consumption to be.
- a quantization factor, a block size of a block includes and encompasses this block with the block size of the quantized coefficient, at least quenzwert a frequency, wherein the quantized coefficient a Amplitude of the at least one frequency value, and / or the temporal prediction mode, in particular an INTRA prediction mode, an INTER prediction mode and / or a RESIDUAL-Präditionsmodus, taken into account.
- Correction values overbased coefficients can be adjusted by using one or more of these characteristic parameters optimally coding properties of the quanti ⁇ .
- an estimation function is dependent on at least one characteristic
- the correction value is determined using a method other than the estimation function, an estimated value for the correction value is generated, a difference value of correction value and estimated value is generated, and the generated difference value is transmitted from an encoding device of the compression method to a decoding device of the compression method If an excess amount of data for signaling the correction values is kept low, since only difference values are transmitted.
- a plurality of correction values or difference values are ⁇ by an encoding device of the compression method to a decoding device of the compression method received.
- a correction value assigned to a characteristic parameter but depending on the value of the characteristic parameter, an optimized correction value can be selected.
- a correction value is such ermit ⁇ telt that the quantization of a coefficient set of quantized coefficients using the characteristic parameters of the selection is minimized, the respective inverted se quantization by an addition of the correction value and the quantized coefficients the coefficient group takes place.
- a nearly optimal correction value can be determined.
- the correction value for a coefficient group in depen ⁇ be adjusted dependence of already processed quantized coefficients and the quantization error is further reduced.
- a video coding method in particular according to the H.264 standard, is selected as the compression method, since it is precisely in the video coding method that an application of the method according to the invention of a clear picture quality improvement can be achieved.
- the coefficient group is the quantized Koeffi ⁇ coefficient of at least one image block, or at least one group of image blocks, or at least an image taken.
- Video coding method predetermined coding structures, a coefficient group and the associated optimal correction ⁇ value determined and a respective quantization error minimized.
- the invention further relates to a device for reducing a quantization error in an inverse quantization of a quantized coefficient in the context of a compression method, wherein the quantization error describes a difference between the quantized coefficient after the inverse quantization and an associated coefficient before quantization, in which a correction value unit is designed such that at least one of the compression of the quantized coefficients charac- diagrammatic parameter is selected, wherein at least two cha ⁇ teristic parameters are selected, if one of the characteristic parameters corresponding to a temporal prediction mode, depending on the selected at least one characteristic parameter of a Correction value is selected, the inverse quantization is performed after addition of the correction value and the quantized coefficient.
- a correction value unit is designed such that at least one of the compression of the quantized coefficients charac- diagrammatic parameter is selected, wherein at least two cha ⁇ teristic parameters are selected, if one of the characteristic parameters corresponding to a temporal prediction mode, depending on the selected at least one characteristic parameter of a Correction value is selected, the
- the device is equipped with a Cor ⁇ rekturwert analyses, which is configured such that, due to the selection of at least one characteristic patent rameters a correction value is determined such that the quantization of a coefficient set of quantized coefficients using the characteristic parameters of the selection is minimized, wherein the respective inverse quantization ⁇ tion is carried out by an addition of the correction value and the original quantized coefficients of the coefficient group.
- a minima ⁇ ler correction value is achieved by formation of a coefficient group, where the quantized coefficients of the coefficient group has been encoded with the same characteristic parameters, is to be determined as that quantized coefficient for the optimum correction value.
- By forming the coefficient group it is achieved that one or more correction values are generated which represent the optimum correction values for all the quantized coefficients within the coefficient group.
- Figure 1 shows a schematic structure of a JPEG encoder (prior art).
- FIG. 2 is a schematic structure of a JPEG decoder (prior art).
- FIG. 3 shows a modified structure of a JPEG encoder for generating correction values for the method according to the invention
- FIG. 4 representation of optimized correction values as a function of quantization stages
- FIG. 5 shows a modified structure of a JPEG decoder for carrying out the method according to the invention taking account of correction values
- FIG. 6 depicts optimum correction values as a function of quantized coefficients and quantization factors
- FIG. 7 shows normalized difference values as a function of quantization factors
- FIG. 8 depicts a respective image quality of two coded video signals
- FIG. 9 system with video server and a mobile terminal for carrying out the method according to the invention.
- FIG. 3 is based on FIG. 1, wherein a correction value unit KWE has been added.
- the ⁇ generated from an image IM compressed data stored in a first file JDSl.
- a quantized coefficient Z is generated for each coefficient X.
- the quantization factor Q, the coefficients X and the quantized coefficients Z are supplied to the correction value unit KWE as coefficient set.
- the quantization factor Q used for each first quantization error QF1 is stored.
- the correction value unit KWE determines a correction value
- the sliding ⁇ monitoring (6) is minimized as a function of the correction value, derived for example by the correction value KW and set to zero, ie
- the quantization error QF for the correction value KW is to be determined for more than one coefficient set, the derived quantization errors QF 1 of the respective coefficient sets are added together and the addition result is set to zero. Those quantized coefficients which are taken into account in this calculation are combined in a group of coefficients G.
- the quantized coefficients Ko ⁇ efficient of the group G is the reference 200506343
- the optimal correction value KW 2/3.
- the optimal Kor have been registered ⁇ compensation values KW.
- the determined optimum correction values KW will be stored, for example, in the first file JDS1. This or these correction values KW are referred to as optimal as they reach minimum quantization error QF ⁇ approximately according to the equation (7).
- the present invention is not limited to the use of the equations (6) and (7) for determining the optimal correction value KW.
- any optimization method such as trying different correction values KW, can be used for this.
- different optimization methods can be used for the generation of the correction values KW for each quantization stage Q. This applies analogously to the case in which other characteristic parameters P than the quantization stage Q are used.
- an optimum correction value KW for all quantization factors Q are determined in which the derived quantization QF added together 1 of the respective sets of coefficients and the resulting ⁇ nis is set to zero.
- the correction value KW 6/17.
- FIG. 4 shows by way of example a function of optimum correction values KW plotted against the quantization factor Q. This function is indicated in FIG. 4 by the reference character KW (Q).
- FIG. 5 shows a decoding unit which is based on the decoding unit according to FIG. In this case, be read zusharm ⁇ Lich to Figure 2, one or more correction values KW from the first data JDSl in decoding the coded picture data, for example by a correction value analysis unit KWA. 200506343
- the associated correction value KW is added to the quantized coefficient Z. Thereafter, the inverse quantization takes place.
- the value KW * Q can be added to the reconstructed coefficient Y only after the inverse quantization.
- the reconstructed coefficient Y Z * Q + KW * Q.
- the optimum correction values KW have been determined taking into account the respective quantization factor Q. This is based on the knowledge that a frequency distribution of the coefficients X changes depending on the quantization factor Q and thus a minimum quantization error QF can be achieved by means of a correction value KW associated with each quantization factor Q.
- the quantization factor Q represents only one of the possible characteristic parameters P that can be taken into account when determining the optimal correction value (s) KW.
- the quantized coefficient Z is taken into account as a characteristic parameter P.
- a respective coefficient group G can be formed for one or more quantized coefficients Z and an optimum correction value KW can be generated for each coefficient group G.
- the quantized coefficients Z in the value range from 0 to 9 and in the range from 10 to 20 are divided into two separate coefficient groups G and an optimum correction value KW is calculated for each coefficient group G.
- At least two characteristic parameters P are used to form the coefficient group G and thus to determine the optimum correction value KW.
- the method according to the invention is used, for example, in a video coding method.
- the following characteristic parameters P can additionally and / or alternatively be used to determine the optimum correction value KW.
- image data is divided into image blocks having a block size BG of eg 4x4, 8x8 or 16x16 pixels.
- BG block size of the image block BB
- an optimal correction value KW for the quantized coefficient Z can be found.
- Frequency value FW In a block-based compression method, such as JPEG or H.264, image blocks of a size of 8 ⁇ 8 pixels are transformed from the local to the frequency range or from the frequency to the local range. In Figure 1 this is achieved by the dis crete cosine transform ⁇ FDCT or the inverse cosine transformation IDCT accomplished.
- each coefficient X and thus also the associated quantized coefficient Z within the transformed image block represents a specific two-dimensional frequency value FW, wherein the 200506343
- Amplitude of the frequency value FW the coefficient X ent ⁇ speaks.
- a corresponding optimal correction value KW is created for each frequency value FW.
- a group G of quantized coefficients Z can be formed from a plurality of frequency values FW.
- a separate optimum correction value KW is generated for the frequency value FW with a DC component DC and for all other frequency values FW, ie for the non-DC components AC.
- a temporal prediction can be used to increase the compression rate.
- the skilled man ⁇ are such temporal predictions as INTRA prediction mode INTRA, INTER INTER prediction mode and RESIDUAL prediction RES example, [1] or [2] known, so that approached their functioning do not further ⁇ .
- the coefficient X is encoded by means of the INTER prediction mode.
- the characteristic parameter P is the inter-prediction mode.
- a characteristic parameter P e.g. the temporal prediction mode PM, not alone but only in combination with another characteristic parameter P, e.g. the block size BG, is used.
- the correction value KW has been used in the inverse quantization in the decoding unit.
- Many compression techniques such as the video coding standard MPEG-4, decode the encoded image data under the INTER-prediction on time vorange ⁇ gangene images to have access.
- an inverse quantization of the quantized coefficient Z is also performed.
- Ver also the invention can travel ⁇ be used, but it should be ensured that the encoding unit and the decoding unit use the same correction value KW at the verses in ⁇ quantization of a specific quantized coefficients Z. Otherwise, the encoding unit and the decoding unit do not operate synchronously and the commisioning method supplies erroneously decoded pictures IM 1 .
- each coefficient group G quantized coefficients Zn considered with at least one characteristic parameter P.
- only certain quantized coefficients can be considered to form one or more coefficient groups G.
- the quantized coefficients Zn of at least one image block BB, and / or at least one group GOB of image blocks and / or at least one image IM can be considered.
- a time at which this is done depends on a number of already quantized coefficients Z or is performed at fixed predetermined times TO.
- an estimation function SF is used to select the correction value KW.
- This estimator SF has the task that a correction value KW can be obtained from this estimator SF.
- To determine this estimation function SF is offline, for example by means of a o- the plurality of test sequences with the block sizes BG 4x4, 8x8 and 16x16 coded and for each of these block sizes a Koeffi ⁇ zientenement G from the quantized coefficient Z of this 200506343
- This estimation function SF is stored in the encoding and / or decoding unit and can be used for the encoding and / or decoding for the selection of a correction value KW when carrying out a compression method. It is advantageous here that the determination of the estimation function SF can be carried out offline.
- the use of the estimation function SF is advantageous since, in comparison to the exemplary embodiments according to FIGS. 3 and 5, a transmission of the correction value (s) KW can be omitted since both the encoding and decoding unit can obtain the correction value KW from the estimation function SF ,
- the estimator SF can also be formed as a function of several characteristic parameters P.
- a statistical function such as a Laplace function, which represents an approximation for a computationally determined estimator.
- FIG. 4 shows an estimation function SF (Q) as a function of the quantization factor Q.
- this estimation function SF (Q) differs from the optimum correction values KW for larger quantization factors Q.
- the estimation function SF and / or the difference value UW can be determined in the correction value unit KWE.
- the estimation function SF and / or the difference value UW can be processed with the aid of the correction value analysis unit KWA.
- a larger amount of data must be transmitted or stored. This can be done, for example, in tabular form.
- a piecewise linearization of the estimation function and / or the function of the difference values can be carried out before transmission or storage. This is shown in FIG. 7 with the dashed function, labeled UW 1 1 (Q).
- the determination of the piecewise linearization or other methods for linearization can be found, for example, in Chapters 3 and 4 of the document [3].
- Figure 5 represents an encoding unit JC, which, such as JPEG, other compression methods such as also can realize a video coding method or audio coding instead of a Schmcodierver ⁇ driving.
- JC encoding unit
- a co-founded video sequence in kilobits per second PSNR (dB) of the encoded video sequence is kbit / s quality of an image displayed ⁇ .
- the solid line function represents a coded video signal VSO of a video sequence without using the method according to the invention.
- the dashed line function shows an optimized coded video signal VSM of the same video sequence as the coded video signal
- VSO wherein the inventive method has been used. Since the optimized encoded video signal VSM is above the encoded video signal VSO, shows the optimized co ⁇ ied video signal VSM at the same data rate BR an improved image quality. This improvement increases here at hö ⁇ heren data rates.
- FIG. 9 shows a video streaming system which comprises a video server VS and a mobile terminal MG, in particular according to the GSM standard (GSM-Global System for Mobile Communications).
- the video server VS used in addition to a Kompressionsal ⁇ rithm process of the invention in the preparation of optimized encoded video signal VSM, with the inventive method by means of the correction value unit KWE is performed.
- the optimized coded video signal VSM comprises, in addition to compressed image data, one or more correction values KW.
- the optimized, encoded video signal VSM will carry over a transmission channel UB to the mobile terminal MG over ⁇ .
- This transmission channel UB is, for example, by means of a wired network, in particular by means of the LAN standard (LAN Local Area Network, or by means of a wireless network, in particular by means of the WLAN standard (WLAN Wireless Local Area Network) or the UMTS standard 200506343
- LAN standard LAN Local Area Network
- WLAN standard WLAN Wireless Local Area Network
- the mobile terminal decodes the received MG opti mized ⁇ encoded video signal VSM and sets the correction value ⁇ analysis unit KWA to implement the encryption of the invention driving a.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
- Image Processing (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP06725548A EP1869890B1 (de) | 2005-04-12 | 2006-04-04 | Verfahren und vorrichtung zur reduktion eines quantisierungsfehlers |
JP2008505863A JP4956527B2 (ja) | 2005-04-12 | 2006-04-04 | 量子化誤差を低減する方法および装置 |
US11/918,399 US8559502B2 (en) | 2005-04-12 | 2006-04-04 | Method and device for minimizing a quantization error |
ES06725548T ES2401694T3 (es) | 2005-04-12 | 2006-04-04 | Procedimiento y dispositivo para reducir un error de cuantificación |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102005016858.2 | 2005-04-12 | ||
DE102005016858A DE102005016858A1 (de) | 2005-04-12 | 2005-04-12 | Verfahren und Vorrichtung zur Reduktion eines Quantisierungsfehlers |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2006108780A1 true WO2006108780A1 (de) | 2006-10-19 |
Family
ID=36763325
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2006/061312 WO2006108780A1 (de) | 2005-04-12 | 2006-04-04 | Verfahren und vorrichtung zur reduktion eines quantisierungsfehlers |
Country Status (6)
Country | Link |
---|---|
US (1) | US8559502B2 (de) |
EP (1) | EP1869890B1 (de) |
JP (1) | JP4956527B2 (de) |
DE (1) | DE102005016858A1 (de) |
ES (1) | ES2401694T3 (de) |
WO (1) | WO2006108780A1 (de) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140117754A (ko) * | 2013-03-26 | 2014-10-08 | 삼성전자주식회사 | 움직임 벡터 처리 장치 및 처리 방법 |
JP6417815B2 (ja) * | 2014-09-19 | 2018-11-07 | 富士ゼロックス株式会社 | 情報処理装置及び情報処理プログラム |
Family Cites Families (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0793583B2 (ja) * | 1986-03-17 | 1995-10-09 | キヤノン株式会社 | 予測符号化装置 |
JPH0629934A (ja) * | 1992-07-10 | 1994-02-04 | Nippon Hoso Kyokai <Nhk> | 適応差分符号化伝送方法 |
US6414992B1 (en) * | 1999-01-27 | 2002-07-02 | Sun Microsystems, Inc. | Optimal encoding of motion compensated video |
US7088780B2 (en) * | 2001-05-11 | 2006-08-08 | Mitsubishi Electric Research Labs, Inc. | Video transcoder with drift compensation |
US6950463B2 (en) * | 2001-06-13 | 2005-09-27 | Microsoft Corporation | Non-compensated transcoding of a video stream |
KR100584552B1 (ko) * | 2003-01-14 | 2006-05-30 | 삼성전자주식회사 | 동영상 부호화 및 복호화 방법과 그 장치 |
EP1569458A1 (de) * | 2004-02-12 | 2005-08-31 | Matsushita Electric Industrial Co., Ltd. | Kodierung und Dekodierung von Videobildern mit nichtlinearer Quantisierung |
US7684632B2 (en) * | 2005-05-16 | 2010-03-23 | Hewlett-Packard Development Company, L.P. | Estimating image compression quantization parameter values |
-
2005
- 2005-04-12 DE DE102005016858A patent/DE102005016858A1/de not_active Withdrawn
-
2006
- 2006-04-04 US US11/918,399 patent/US8559502B2/en not_active Expired - Fee Related
- 2006-04-04 EP EP06725548A patent/EP1869890B1/de not_active Not-in-force
- 2006-04-04 JP JP2008505863A patent/JP4956527B2/ja not_active Expired - Fee Related
- 2006-04-04 WO PCT/EP2006/061312 patent/WO2006108780A1/de active Application Filing
- 2006-04-04 ES ES06725548T patent/ES2401694T3/es active Active
Non-Patent Citations (3)
Title |
---|
KWANG-DEOK SEO ET AL: "Laplacian model-based inverse quantization for DCT-based image codec system", CIRCUITS AND SYSTEMS, 2004. ISCAS '04. PROCEEDINGS OF THE 2004 INTERNATIONAL SYMPOSIUM ON VANCOUVER, BC, CANADA 23-26 MAY 2004, PISCATAWAY, NJ, USA,IEEE, US, 23 May 2004 (2004-05-23), pages III - 881, XP010719403, ISBN: 0-7803-8251-X * |
PRICE J R ET AL: "Biased reconstruction for JPEG decoding", IEEE SIGNAL PROCESSING LETTERS, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 6, no. 12, December 1999 (1999-12-01), pages 297 - 299, XP011060178, ISSN: 1070-9908 * |
SORIAL H ET AL: "Estimating Laplacian parameters of DCT coefficients for requantization in the transcoding of MPEG-2 video", IMAGE PROCESSING, 2000. PROCEEDINGS. 2000 INTERNATIONAL CONFERENCE ON SEPTEMBER 10-13, 2000, PISCATAWAY, NJ, USA,IEEE, 10 September 2000 (2000-09-10), pages 956 - 959, XP010530775, ISBN: 0-7803-6297-7 * |
Also Published As
Publication number | Publication date |
---|---|
DE102005016858A1 (de) | 2006-10-19 |
US20090067492A1 (en) | 2009-03-12 |
EP1869890A1 (de) | 2007-12-26 |
US8559502B2 (en) | 2013-10-15 |
JP4956527B2 (ja) | 2012-06-20 |
ES2401694T3 (es) | 2013-04-23 |
JP2008537866A (ja) | 2008-09-25 |
EP1869890B1 (de) | 2013-02-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
DE69805583T2 (de) | Quantisierungsmatrix für die codierung von stand- und bewegtbildern | |
DE69524141T2 (de) | Adaptiver Bildkodierer mit dynamischer Bitzuteilung | |
DE102016125379B4 (de) | Bewegungsvektoraufteilung des letzten Frames | |
DE102016125353A1 (de) | Kodierratensteuerung von echtzeitvideos unter verwendung einer dynamischen auflösungsumschaltung | |
WO2012095487A1 (de) | Kompression und dekompression von referenzbildern in einem videokoder | |
DE102019103346A1 (de) | Ssim-basierte rate-distortion-optimierung für eine verbesserte wahrnehmbare qualität bei videos | |
EP1774790B1 (de) | Verfahren und vorrichtung zum codieren und decodieren | |
DE10343220B3 (de) | Verfahren und Vorrichtung zur Transcodierung eines Datenstroms, der ein oder mehrere codierte digitalisierte Bilder umfasst | |
DE3889939T2 (de) | Verfahren und Mittel zur Kodierung und Dekodierung eines Bildsignals. | |
EP1425920B1 (de) | Verfahren zur videocodierung und computerprogrammprodukt | |
WO2006108780A1 (de) | Verfahren und vorrichtung zur reduktion eines quantisierungsfehlers | |
EP1285537B1 (de) | Verfahren und eine anordnung zur codierung bzw. decodierung einer folge von bildern | |
WO2002054774A2 (de) | Optimale snr-skalierbare videocodierung | |
DE10121259C2 (de) | Optimale SNR-skalierbare Videocodierung | |
DE102020207711B3 (de) | Verfahren und vorrichtungen zur codierung von bilddaten | |
DE10219640A1 (de) | Verfahren zum Codieren und Decodieren von Videosequenzen und Computerprogrammprodukt | |
DE102004063902B4 (de) | Computerprogramm mit einem Verfahren zum Verarbeiten einer Gruppe von Bildern und mit einem Verfahren zum Verarbeiten eines Basisbildes und eines oder mehrerer Erweiterungsbilder | |
WO2003026310A2 (de) | Effiziente videocodierung für skalierbare simul-cast-speicherung und -übertragung | |
DE102004011421B4 (de) | Vorrichtung und Verfahren zum Erzeugen eines skalierten Datenstroms | |
DE102004011422B4 (de) | Vorrichtung und Verfahren zum Verarbeiten einer Gruppe von Bildern und Vorrichtung und Verfahren zum Verarbeiten eines Basisbildes und eines oder mehrerer Erweiterungsbilder | |
EP0284962A2 (de) | Verfahren zur Bilddatenreduktion für digitale Fernsehsignale | |
EP1157557A1 (de) | Verfahren und anordnung zur transformation eines bildbereichs | |
WO2003026271A2 (de) | Verfahren zum kodieren und dekodieren von videosequenzen und computerprogrammprodukt | |
WO2001063939A1 (de) | Verfahren und vorrichtung zum codieren eines videosignals und eines sprachsignals, computerlesbares speichermedium sowie computerprogramm-element | |
DE10243568A1 (de) | Verfahren zur skalierbaren Videocodierung eines Videobildsignals sowie ein zugehöriger Codec |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2006725548 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008505863 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: DE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
NENP | Non-entry into the national phase |
Ref country code: RU |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: RU |
|
WWP | Wipo information: published in national office |
Ref document number: 2006725548 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 11918399 Country of ref document: US |