CN107071417B - A kind of intra-frame prediction method for Video coding - Google Patents

A kind of intra-frame prediction method for Video coding Download PDF

Info

Publication number
CN107071417B
CN107071417B CN201710228419.6A CN201710228419A CN107071417B CN 107071417 B CN107071417 B CN 107071417B CN 201710228419 A CN201710228419 A CN 201710228419A CN 107071417 B CN107071417 B CN 107071417B
Authority
CN
China
Prior art keywords
pixel
formula
following formula
block
pixel values
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201710228419.6A
Other languages
Chinese (zh)
Other versions
CN107071417A (en
Inventor
朱策
毛敏
刘宇洋
宋世昶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN201710228419.6A priority Critical patent/CN107071417B/en
Publication of CN107071417A publication Critical patent/CN107071417A/en
Application granted granted Critical
Publication of CN107071417B publication Critical patent/CN107071417B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • H04N19/102—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
    • H04N19/117—Filters, e.g. for pre-processing or post-processing
    • 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
    • H04N19/134—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
    • H04N19/167—Position within a video image, e.g. region of interest [ROI]
    • 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
    • H04N19/80—Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

The invention belongs to technical field of video coding, particularly relate to a kind of intra-frame prediction method for Video coding.The method comprise the steps that generating filtered reference pixel using the unfiltered reference pixel of multiple lines and multiple rows, generating predicted value using filtered reference pixel, smoothly generate new predicted value using the unfiltered reference pixel of multiple lines and multiple rows and predicted value.Beneficial effects of the present invention are relatively existing PDPC technology, the present invention can more efficiently, robustly carry out intraframe coding.

Description

A kind of intra-frame prediction method for Video coding
Technical field
The invention belongs to technical field of video coding, particularly relate to a kind of intra prediction side for Video coding Method.
Background technique
In recent years, colourful video content is all generating all the time, and number of videos is even more that exponentially grade is explosive Increase.It is counted according to YouTube, the amount of video that user uploads per minute was up to more than 300 hours；AT&T Labs's prediction, the year two thousand twenty sound Video data flow will account for the 80% of new flow increasing, there is some evidence that, video data has become the big data in big data, I Be in the multimedia era surrounded by magnanimity audio, video data.Transmission band occupation rate is reduced in order to save memory space, It generally requires and does video compression coding processing.
Intraframe coding is one of video encoding standard coding mode.The reference pixel of traditional intraframe coding merely with One column pixel of a line apart from the nearest reconstructed block of current block to be encoded, by selecting a kind of prediction angle mode to carry out linear prediction It obtains.
This prediction technique single-row using uniline has been proved to be very effective small size encoding block, but this Kind of method is to the effect of large scale encoding block and bad, and the position of current block to be encoded is different, the prediction that encoder needs Pixel filter strategy is also not quite similar.For example, more low intensive smooth relative to being needed far from the upper left pixel of block to be encoded. In addition, since encoding block content is different, encoder needs different filtering strategies even for the block of same size.
In the past period, researchers are proposed using recursion filter smoothing prediction value, experiments have shown that can be with Improve code efficiency.
W.-J.Han and J.Min (" Improved video compression efficiency through flexible unit representation and corresponding extension of coding tools,” IEEE Trans.Circuits Syst.Video Technol.,vol.20,no.12,pp.1709–1720,Dec.2010.) And Y.Chen and J.Han, (" A recursive extrapolation approach to intra prediction in video coding,”in Proc.IEEE Int.Conf.Acoustics Speech Signal Process., Vancouver, BC, Canada, May 2013, pp.1734-1738.), it is to carry out smoothing prediction value using recursion filter group.Though Right recursive method can to a certain extent carry out smoothly prediction pixel, and the experimental results showed that recursion filter can mention Coding efficiency is risen, however, recursion filter is not particularly suited for parallel computation, and does not consider the variation of signal entirety.
Summary of the invention
In order to preferably utilize block to be encoded inner vein feature, the present invention propose it is a kind of based on the position of multiple lines and multiple rows according to Intra-frame prediction method (the Enhanced Position dependent intra prediction for relying prediction combined combination,EPDPC)。
The technical scheme is that
A kind of intra-frame prediction method for Video coding, which comprises the following steps:
S1, filtered reference pixel is generated:
Filtered reference pixel S is obtained by following formula 1:
S=Ak·R+(E-Ak)·(Hk* R) (formula 1)
In formula 1, HkFilter parameter under the coding block size, R are the multiple lines and multiple rows for the current block to be encoded chosen Reference pixel, * are convolution operator, AkFor the weighted factor of filter output；
S2, the predicted pixel values for calculating present encoding block:
Coding block size is set as M × N, to the prediction pixel P of each pixel (x, y) in encoding blockx,yIt carries out pre- It surveys, wherein N >=x >=1, M >=y >=1；Assuming that Sx,0With S0,yBe respectively in the filtered reference pixel S of current block to be encoded near The one column pixel of a line of nearly current block to be encoded, then the generation method of predicted pixel values include:
Under plane (Planar) prediction mode:
By prediction pixel Px,yIt is defined as the average value of horizontal, vertical both direction predicted value, then passes through following formula 2, public affairs Formula 3 and formula 4 obtain predicted pixel values:
Under DC component (DC) prediction mode:
For chromatic component and greater than the luminance component of 16x16, predicted pixel values are obtained by following formula 5:
Other situations then pass through following formula 6 and obtain:
Under angle mode:
For giving vertical class angle mode M, shape is mapped as using the reference pixel that projected pixel method is needed to use Formula is denoted as Ref, following formula 7 and formula 8:
It calculates current pixel and corresponds to position of the reference pixel in, be denoted as pos, following formula 9:
Pos=(yoffset [M]) > > 5 (formula 9)
The weighted factor w that current pixel corresponds to reference pixel is calculated, following formula 10:
W=(yoffset [M]) &31 (formula 10)
The predicted pixel values of current pixel are calculated by following formula 11:
Px,y=((32-w) Ref [pos]+wRef [pos+1]+16) > > 5 (formula 11)
Wherein, offset [M] is angle commutation factor, and Round () is the operation function that rounds up；
S3, new predicted value is smoothly generated according to the unfiltered reference pixel of multiple lines and multiple rows and predicted value:
According to the predicted value P obtained in step S2x,y, obtained by following formula 12- formula 16 final current to be encoded The predicted pixel values of block:
Wherein,It is the Prediction Parameters of storage, for 16 × 16 block, d=1 is set, for more D=2 is arranged in big block；bx,yIt is normalization factor, following formula 16:
It is the weighting weight for being to rely on horizontal position, upright position respectively,It is one group for the upper left corner The weighting weight of position,Dependent on horizontal position, upright position, upper left position convolution mask,Be to Lower rounding operation symbol, > > left shift operator.
The total technical solution of this law, according to prediction angle mode, coded block size and the location of pixels of encoding block, using more The unfiltered reference pixel of row multiple row predicts the predicted value of current block to be encoded, and with after obtained predicted value and filtering (smooth) The unfiltered reference pixel weighted array of multiple lines and multiple rows obtains final predicted value.Realize efficiently intraframe coding.
The beneficial effects of the present invention are: compared with the prior art, the present invention can more efficiently, robustly carry out compiling in frame Code.
Detailed description of the invention
Fig. 1 is that two rows two arrange EPDPC schematic diagram in the present invention, wherein (a) current block to be encoded and unfiltered reference image Plain schematic diagram is (b) current block to be encoded and filtered reference pixel schematic diagram；
Fig. 2 is angle model prediction direction schematic diagram in frame, and wherein each arrow represents a prediction direction.
Specific embodiment
With reference to the accompanying drawings and examples, the technical schemes of the invention are described in detail:
Embodiment
This example is that the validity of the method for the invention is verified with the EPDPC of two rows two column.
The method of this example includes:
A quasi- Gaussian filter is constructed using formula (1), by the upper unfiltered column of two row two ginseng of current block to be encoded It examines pixel Unfiltered Refline2 and Unfiltered Refline1 filtering (Fig. 1 (a)) and generates (Fig. 1 (b)) filtering Reference pixel Filtered Refline1:
S=Ak·R+(E-Ak)·(Hk* R) (formula 1)
Step 2, according to fig. 2 middle angle mode calculate current compile with filtered reference pixel Filtered Refline1 Prediction pixel Prediction (the P of code block pixel (x, y)x,y)；
Step 3, with Unfiltered Refline2 and Unfiltered Refline1 and PredictionP (x, y) Filtering generates FPrediction (FPx,y).P is obtained by the method for step 2x,yAfterwards, b is calculated according to formula (16)x,y:
Then the predicted pixel values FP of final current block is calculated using formula (15)x,y:
Parameter in formula (15) can utilize formula (12), and formula (13) and formula (14) are even if obtain；Wherein, for 16 D=1 is arranged in × 16 block, and for bigger block, d=2 is arranged；And it is arrangedIt is one group of convolution mask.
Table 3 is the coding result that embodiment obtains:
The coding result that 3 embodiment of table obtains
Sequence Y U V
Traffic - 0.1% 0.0% 0.0%
PeopleOnStreet - 0.2% - 0.3% - 0.2%
Nebuta - 0.2% - 0.4% - 0.2%
SteamLocomotive - 0.2% - 0.8% - 0.1%
ParkScene - 0.2% - 0.1% - 0.1%
Cactus - 0.1% - 0.1% - 0.2%
BasketballPass - 0.1% - 0.1% - 0.4%
BQSquare - 0.1% 0.1% - 0.1%
BlowingBubbles - 0.1% - 0.2% - 0.1%
RaceHorses - 0.1% - 0.2% - 0.1%
FourPeople - 0.2% - 0.2% - 0.3%
Johnny - 0.1% - 0.1% - 0.1%
KristenAndSara - 0.1% 0.0% 0.0%
SlideShow - 0.2% - 0.2% - 0.6%
(average) Average - 0.1% - 0.2% - 0.2%
The coding efficiency shown in the table 3 it can be concluded that, the encoding efficiency of tri- components of YUV is improved.

Claims (1)

1. a kind of intra-frame prediction method for Video coding, which comprises the following steps:
S1, filtered reference pixel is generated:
Filtered reference pixel S is obtained by following formula 1:
S=Ak·R+(E-Ak)·(Hk* R), formula 1；
In formula 1, HkFilter parameter under the coding block size, R are the multiple lines and multiple rows reference images for the current block to be encoded chosen Element, * are convolution operator, AkFor the weighted factor of filter output；
S2, the predicted pixel values for calculating present encoding block:
Coding block size is set as M × N, to the prediction pixel P of each pixel (x, y) in encoding blockx,yIt is predicted, Wherein, N >=x >=1, M >=y >=1；Assuming that Sx,0With S0,yBe respectively in the filtered reference pixel S of current block to be encoded near The one column pixel of a line of current block to be encoded, then the generation method of predicted pixel values include:
Under plane prediction mode:
By prediction pixel Px,yIt is defined as the average value of horizontal, vertical both direction predicted pixel values, then passes through following formula 2, public affairs Formula 3 and formula 4 obtain predicted pixel values:
Under DC component prediction mode:
For chromatic component and greater than the luminance component of 16x16, predicted pixel values are obtained by following formula 5:
Other situations then pass through following formula 6 and obtain:
Under angle mode:
For giving vertical class angle mode M, it is mapped as form using the reference pixel that projected pixel method is needed to use, is remembered For Ref, following formula 7 and formula 8:
It calculates current pixel and corresponds to position of the reference pixel in, be denoted as pos, following formula 9:
Pos=(yoffset [M]) > > 5, formula 9；
The weighted factor w that current pixel corresponds to reference pixel is calculated, following formula 10:
W=(yoffset [M]) &31, formula 10；
The predicted pixel values of current pixel are calculated by following formula 11:
Px,y=((32-w) Ref [pos]+wRef [pos+1]+16) > > 5, formula 11；
Wherein, offset [M] is angle commutation factor, and Round () is the operation function that rounds up；
S3, new predicted pixel values are smoothly generated according to the unfiltered reference pixel of multiple lines and multiple rows and predicted pixel values:
According to the predicted pixel values P obtained in step S2x,y, obtained by following formula 12- formula 16 final current to be encoded The predicted pixel values of block:
Wherein,It is the Prediction Parameters of storage, for 16 × 16 block, d=1 is set, for bigger D=2 is arranged in block；bx,yIt is normalization factor, following formula 16:
It is the weighting weight for being to rely on horizontal position, upright position respectively,It is one group for upper left position Weight is weighted,Dependent on horizontal position, upright position, upper left position convolution mask,It is to be rounded downwards Operator, > > left shift operator.
CN201710228419.6A 2017-04-10 2017-04-10 A kind of intra-frame prediction method for Video coding Active CN107071417B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710228419.6A CN107071417B (en) 2017-04-10 2017-04-10 A kind of intra-frame prediction method for Video coding

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710228419.6A CN107071417B (en) 2017-04-10 2017-04-10 A kind of intra-frame prediction method for Video coding

Publications (2)

Publication Number Publication Date
CN107071417A CN107071417A (en) 2017-08-18
CN107071417B true CN107071417B (en) 2019-07-02

Family

ID=59601723

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710228419.6A Active CN107071417B (en) 2017-04-10 2017-04-10 A kind of intra-frame prediction method for Video coding

Country Status (1)

Country Link
CN (1) CN107071417B (en)

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10856010B2 (en) 2017-09-08 2020-12-01 FG Innovation Company Limited Device and method for coding video data based on multiple reference lines
US10965941B2 (en) 2017-10-09 2021-03-30 Qualcomm Incorporated Position-dependent prediction combinations in video coding
CN110858281B (en) * 2018-08-22 2022-10-04 浙江宇视科技有限公司 Image processing method, image processing device, electronic eye and storage medium
JP7401542B2 (en) 2018-11-26 2023-12-19 華為技術有限公司 How to intra-predict blocks of pictures
JP7277586B2 (en) * 2018-12-21 2023-05-19 ホアウェイ・テクノロジーズ・カンパニー・リミテッド Method and apparatus for mode and size dependent block level limiting
CN116567207B (en) * 2018-12-29 2024-02-13 华为技术有限公司 Method and apparatus for intra prediction
CN112514381A (en) 2019-06-25 2021-03-16 Oppo广东移动通信有限公司 Image encoding and decoding method, encoder, decoder, and storage medium
WO2020258001A1 (en) * 2019-06-25 2020-12-30 Oppo广东移动通信有限公司 Luminance component prediction method, encoder, decoder and storage medium
CN113826403A (en) 2019-06-25 2021-12-21 Oppo广东移动通信有限公司 Information processing method and device, equipment and storage medium
CN111787334B (en) * 2020-05-29 2021-09-14 浙江大华技术股份有限公司 Filtering method, filter and device for intra-frame prediction
CN114363632B (en) * 2021-12-10 2023-05-16 浙江大华技术股份有限公司 Intra-frame prediction method, encoding and decoding method, encoder and decoder, system, electronic device, and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102934446A (en) * 2010-04-05 2013-02-13 三星电子株式会社 Determining intra prediction mode of image coding unit and image decoding unit
CN103747250A (en) * 2013-12-28 2014-04-23 华中科技大学 Method for 4*4 sub-macroblock parallel intraframe prediction in H.264/AVC
CN105516728A (en) * 2015-12-15 2016-04-20 华中科技大学 Parallel intra-frame prediction method of 8*8 sub-macroblocks in H.265/HEVC

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102934446A (en) * 2010-04-05 2013-02-13 三星电子株式会社 Determining intra prediction mode of image coding unit and image decoding unit
CN103747250A (en) * 2013-12-28 2014-04-23 华中科技大学 Method for 4*4 sub-macroblock parallel intraframe prediction in H.264/AVC
CN105516728A (en) * 2015-12-15 2016-04-20 华中科技大学 Parallel intra-frame prediction method of 8*8 sub-macroblocks in H.265/HEVC

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
AVS2视频编码码率控制算法;罗敏珂，周益民，钟敏，朱策;《系统工程与电子技术》;20160930;全文 *
H.264编码器预测算法优化研究;徐宁;《中国优秀硕士论文库》;20060228;全文 *

Also Published As

Publication number Publication date
CN107071417A (en) 2017-08-18

Similar Documents

Publication Publication Date Title
CN107071417B (en) A kind of intra-frame prediction method for Video coding
CN101091393B (en) Moving picture encoding method, device using the same
US11838509B2 (en) Video coding method and apparatus
CN101454990B (en) Video compression method
CN102498719B (en) Based on skip mode, image is carried out the method and apparatus of Code And Decode
CN106170093B (en) Intra-frame prediction performance improving coding method
CN101268475B (en) classified filtering for temporal prediction
CN103517069A (en) HEVC intra-frame prediction quick mode selection method based on texture analysis
CN102420988B (en) Multi-view video coding system utilizing visual characteristics
CN102132566B (en) Image processing device and method, dynamic image encoding device and method,dynamic image decoding device and method,and encoding/decoding system and method
US7760964B2 (en) Method and architecture for temporal-spatial deblocking and deflickering with expanded frequency filtering in compressed domain
Deligiannis et al. Side-information-dependent correlation channel estimation in hash-based distributed video coding
CN103179403A (en) Method and apparatus for encoding video, and method and apparatus for decoding video
CN106101707A (en) Block elimination filtering
WO2012096623A1 (en) Deblocking filtering
TW201004368A (en) Low-complexity and high-quality error concealment techniques for video sequence transmissions
CN103119935A (en) Image interpolation method and apparatus
CN108833925B (en) Inter-frame prediction method based on deep neural network
CN105812797B (en) A kind of coding unit selection method and device
CN103096060B (en) The adaptive loop filter method and device of intra-frame image prediction encoding and decoding
CN1089983C (en) Method and apparatus for limiting band of moving-picture signal
CN110115033A (en) Hybrid domain for damaging still image coding cooperates with postfilter
CN103491371A (en) Encoding method, device and equipment based on hierarchy
CN110944211B (en) Interpolation filtering method, device, medium and electronic device for intra-frame prediction
CN104581155A (en) Scenario-analysis-based coding method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant