(19)
(11)EP 2 335 419 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
04.11.2020 Bulletin 2020/45

(21)Application number: 09789245.9

(22)Date of filing:  01.09.2009
(51)International Patent Classification (IPC): 
H04N 19/59(2014.01)
H04N 19/85(2014.01)
(86)International application number:
PCT/US2009/004950
(87)International publication number:
WO 2010/033151 (25.03.2010 Gazette  2010/12)

(54)

METHODS AND APPARATUS FOR VIDEO IMAGING PRUNING

VERFAHREN UND VORRICHTUNG ZUM ZURECHTSCHNEIDEN VON VIDEOBILDMATERIAL

PROCÉDÉS ET APPAREIL POUR LA COUPE D'IMAGES VIDÉO


(84)Designated Contracting States:
AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR

(30)Priority: 18.09.2008 US 98125 P

(43)Date of publication of application:
22.06.2011 Bulletin 2011/25

(73)Proprietor: InterDigital Madison Patent Holdings
75017 Paris (FR)

(72)Inventors:
  • VO, Dung Trung
    La Jolla CA 92037 (US)
  • SOLE, Joel
    La Jolla California 92037 (US)
  • YIN, Peng
    Ithaca NY 14850 (US)
  • ZHENG, Yunfei
    Plainsboro NJ 08536 (US)

(74)Representative: Frese, Beate et al
Frese Patent PartGmbB Patentanwälte
Hüttenallee 237b 47800 Krefeld
Hüttenallee 237b 47800 Krefeld (DE)


(56)References cited: : 
EP-A2- 1 401 211
US-B1- 6 625 322
US-A1- 2002 064 226
  
  • HAAN DE G ET AL: "Deinterlacing - An overview" PROCEEDINGS OF THE IEEE, IEEE. NEW YORK, US, vol. 86, no. 9, 1 September 1998 (1998-09-01), pages 1839-1857, XP011044084 ISSN: 0018-9219
  • EINARSSON T ET AL: "MIXED RESOLUTION VIDEO CODING FOR LOW BIT-RATE CHANNELS" PROCEEDINGS OF THE INTERNATIONAL WORKSHOP ON CODING TECHNIQUESFOR VERY LOW BIT-RATE VIDEO. VLBV, XX, XX, 28 July 1997 (1997-07-28), pages 77-80, XP008028661
  • DUNG TRUNG VO ET AL: "Data pruning-based compression using high order edge-directed interpolation" ACOUSTICS, SPEECH AND SIGNAL PROCESSING, 2009. ICASSP 2009. IEEE INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 19 April 2009 (2009-04-19), pages 997-1000, XP031459400 ISBN: 978-1-4244-2353-8
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

CROSS-REFERENCE TO RELATED APPLICATIONS



[0001] This application claims the benefit of U.S. Provisional Application Serial No. 61/098,125, filed 18 September, 2008.

TECHNICAL FIELD



[0002] The present principles relate generally to video encoding and decoding and, more particularly, to methods and apparatus for video image pruning.

BACKGROUND



[0003] Currently, requests for higher quality video are ever increasing. Video content demands tend to go to higher resolution, higher frame-rate, and higher bit-depth. To combat the bit-rate increase corresponding to high definition (HD) video and other bit rate intensive developments, especially to meet the transmission constraint of network and communication techniques, new technologies to further reduce bit-rate are strongly demanded.

[0004] There are at least two basic approaches to reducing compression bit rate. The first approach involves improving compression technology, and the second approach involves performing some sort of preprocessing prior to compression.

[0005] With respect to the first approach, namely improving compression technology, the progression of the same can be readily seen in the development of the various Moving Picture Experts Group video coding standards, such as, for example, the International Organization for Standardization/International Electrotechnical Commission (ISO/IEC) MPEG-1 Standard, the ISO/IEC MPEG-2 Standard, the ISO/IEC MPEG-4 Standard, and the ISO/IEC MPEG-4 Part 10 Advanced Video Coding (AVC) Standard/International Telecommunication Union, Telecommunication Sector (ITU-T) H.264 Recommendation (hereinafter the "MPEG-4 AVC Standard").

[0006] For most video coding standards, increasing the quantization step size is a means used to reduce the bit-rate. However, this technique can result in severe blocky artifacts and other coding artifacts due to the loss of high frequency details.

[0007] With respect to the second approach, namely performing some sort of preprocessing prior to compression, the goal of such preprocessing is to remove the information that is least important in terms of visual perception, or the information that is able to be recovered after the decoding process without significantly altering the content. This bit rate reduction is also commonly referred to as data pruning. Some common techniques to perform this bit rate reduction through data pruning are the use of low-pass filters and down-sampling (which can be seen as a filtering process), followed by an up-sampling at the decoder. One effect of these schemes is that the decoded and reconstructed video looks a bit blurry since these techniques are designed to eliminate high frequency information in the video in order to reduce the bit-rate.

[0008] As for interpolation which, for example, can be used for the previously described upsampling, a wide range of interpolation methods and schemes have been discussed and developed, beginning with conventional bilinear and bi-cubic interpolation and continuing to sophisticated iterative interpolation methods such as projection onto convex sets (POCS) and non-convex nonlinear partial differential equations.

[0009] To avoid the jerkiness occurring along edges, edge-oriented interpolation methods using a Markov random field and covariance of the low resolution image have been proposed.

[0010] One prior art approach employs a combination of directional filtering and data fusion to estimate the missing high resolution (HR) pixels by a linear minimum mean square error (LMMSE). Another group of interpolation algorithms predict the fine structure of the HR image from its corresponding low resolution (LR) image using different kinds of transforms such as the wavelet transform or the countourlet transform.

[0011] Each of the above methods are suitable for up-sampling the same ratio in both horizontal and vertical directions; that is, in a fixed and regular data grid (i.e., all data points are found in a rectangular grid). However, when interpolation is used along with data pruning, flexibility is desired in order to adapt to the discarded data and to adjust to the varying surroundings of each pixel to achieve the best performance.

SUMMARY



[0012] These and other drawbacks and disadvantages of the prior art are addressed by the present principles, which are directed to methods and apparatus for video image data pruning according to the independent claims.

BRIEF DESCRIPTION OF THE DRAWINGS



[0013] The present principles may be better understood in accordance with the following exemplary figures, in which:

FIG. 1 is a block diagram showing an exemplary system for data pruning and restoration, in accordance with an embodiment of the present principles;

FIG. 2 is a flow diagram showing an exemplary method for data pruning and restoration, in accordance with an embodiment of the present principles;

FIG. 3 is a block diagram showing an exemplary video encoder to which the present principles may be applied, in accordance with an embodiment of the present principles;

FIG. 4 is a block diagram showing an exemplary video decoder to which the present principles may be applied, in accordance with an embodiment of the present principles;

FIG. 5A is a diagram showing model parameters of high-order edge-directed interpolation relating to NEDI-4 and NEDI-8, in accordance with an embodiment of the present principles;

FIG. 5B is a diagram showing model parameters of high-order edge-directed interpolation relating to NEDI-6, in accordance with an embodiment of the present principles;

FIG. 6 is a diagram showing an example of data pruning by dropping arbitrary lines, in accordance with an embodiment of the present principles;

FIG. 7 is a diagram showing an example of data pruning by dropping columns and rows in one frame, in accordance with an embodiment of the present principles;

FIG. 8 is a diagram showing an example of data pruning by dropping the same columns and rows for a whole group of pictures (GOP) and/or scene, in accordance with an embodiment of the present principles;

FIG. 9A is a diagram showing lines for indicating the dropped columns and rows in the first frame of the well-known Akiyo sequence, in accordance with an embodiment of the present principles;

FIG. 9B is a diagram showing the pruned frame corresponding to the example of FIG. 7A, in accordance with an embodiment of the present principles;

FIG. 10 is a block diagram showing an exemplary system for data pruning based compression, in accordance with an embodiment of the present principles;

FIG. 11 is a block diagram showing an exemplary system for a data pruning phase, in accordance with an embodiment of the present principles;

FIG. 12 is a flow diagram showing an exemplary method for optimal data pruning, in accordance with an embodiment of the present principles; and

FIG. 13 is a flow diagram showing an exemplary method for optimal data restoring, in accordance with an embodiment of the present principles.


DETAILED DESCRIPTION



[0014] The present principles are directed to methods and apparatus for video image data pruning.

[0015] The invention concerns pre-processing (215) a picture prior to, and in preparation for, compression by encoding, wherein said pre-processing step comprises selectively removing, based on location of pixels within the spatial domain of the picture, and not in the frequency domain, portions of said picture and leaving the remaining portions of said picture to be encoded using a data pruner having a processor (1200), namely in transmitting side information indicating a location of the removed portions within the picture to a decoder or other device and in selectively removing diagonal line portions of said picture,
wherein the diagonal line portions are neither rows nor columns and encoding the pre-processed picture.

[0016] The present description illustrates the present principles. It will thus be appreciated that those skilled in the art will be able to devise various arrangements that, although not explicitly described or shown herein, embody the present principles and are included within its scope.

[0017] All examples and conditional language recited herein are intended for pedagogical purposes to aid the reader in understanding the present principles and the concepts contributed by the inventor(s) to furthering the art, and are to be construed as being without limitation to such specifically recited examples and conditions.

[0018] Moreover, all statements herein reciting principles, aspects, and embodiments of the present principles, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, it is intended that such equivalents include both currently known equivalents as well as equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.

[0019] Thus, for example, it will be appreciated by those skilled in the art that the block diagrams presented herein represent conceptual views of illustrative circuitry embodying the present principles. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudocode, and the like represent various processes which may be substantially represented in computer readable media and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

[0020] The functions of the various elements shown in the figures may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Moreover, explicit use of the term "processor" or "controller" should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor ("DSP") hardware, read-only memory ("ROM") for storing software, random access memory ("RAM"), and non-volatile storage.

[0021] Other hardware, conventional and/or custom, may also be included. Similarly, any switches shown in the figures are conceptual only. Their function may be carried out through the operation of program logic, through dedicated logic, through the interaction of program control and dedicated logic, or even manually, the particular technique being selectable by the implementer as more specifically understood from the context.

[0022] In the claims hereof, any element expressed as a means for performing a specified function is intended to encompass any way of performing that function including, for example, a) a combination of circuit elements that performs that function or b) software in any form, including, therefore, firmware, microcode or the like, combined with appropriate circuitry for executing that software to perform the function. The present principles as defined by such claims reside in the fact that the functionalities provided by the various recited means are combined and brought together in the manner which the claims call for. It is thus regarded that any means that can provide those functionalities are equivalent to those shown herein.

[0023] Reference in the specification to "one embodiment" or "an embodiment" of the present principles, as well as other variations thereof, means that a particular feature, structure, characteristic, and so forth described in connection with the embodiment is included in at least one embodiment of the present principles. Thus, the appearances of the phrase "in one embodiment" or "in an embodiment", as well any other variations, appearing in various places throughout the specification are not necessarily all referring to the same embodiment.

[0024] It is to be appreciated that the use of any of the following "/", "and/or", and "at least one of', for example, in the cases of "A/B", "A and/or B" and "at least one of A and B", is intended to encompass the selection of the first listed option (A) only, or the selection of the second listed option (B) only, or the selection of both options (A and B). As a further example, in the cases of "A, B, and/or C" and "at least one of A, B, and C", such phrasing is intended to encompass the selection of the first listed option (A) only, or the selection of the second listed option (B) only, or the selection of the third listed option (C) only, or the selection of the first and the second listed options (A and B) only, or the selection of the first and third listed options (A and C) only, or the selection of the second and third listed options (B and C) only, or the selection of all three options (A and B and C). This may be extended, as readily apparent by one of ordinary skill in this and related arts, for as many items listed.

[0025] As used herein, the terms image and picture are used interchangeably, and may refer to any of video pictures that are part of a video sequence and/or still images.

[0026] As noted above, the present principles are directed to methods and apparatus for video image data pruning.

[0027] Moreover, it is to be appreciated that while one or more embodiments of the present principles are described herein with respect to the MPEG-4 AVC standard, the present principles are not limited to solely this standard and, thus, may be utilized with respect to other video coding standards, recommendations, and extensions thereof, including extensions of the MPEG-4 AVC standard, while maintaining the present principles.

[0028] Turning to FIG. 1, an exemplary system for data pruning and restoration is indicated generally by the reference numeral 100. The system 100 includes a data pruner 110, an encoder (or compressor) 120, a decoder (or decompressor) 130, and a data restorer 140. The data pruner 110 performs pre-processing to remove, in the spatial domain, portions (e.g., a line, curve, region, etc.) from a picture prior to compression by encoding. The encoder 120 then codes the picture with the portions removed. The decoder 130 decodes the coded picture with the portions removed. The data restorer 140 performs post-processing to restore, in the spatial domain, the portions to the decoded picture. Any of the data pruner 110, the encoder 120, the decoder 130, and the data restorer 140 may include, for example, one or more processors, memory, and so forth, in order to perform their corresponding functions. For example, at least one or more processors in the data pruner 110 may perform the pre-processing described herein, while one or more processors in the data restorer 140 may perform the post-processing described herein. It is to be appreciated that the data pruner 110 may include other elements which, in some embodiments, may be separate from the data pruner, while in other embodiments may be integrated with the data pruner. For example, in an embodiment, the data pruner 110 (or a data pruning phase) may include a decimator and an interpolator (see, e.g., FIG. 11). In another embodiment, the data pruner 110 (or a data pruning phase) may include a data pruner (as a sub-element that performs the actual removal of the portions), a store or compressor, and an interpolator (see, e.g., FIG. 10). These and other variations of the elements of FIG. 1 are readily contemplated by one of ordinary skill in this and related arts, while maintaining the present principles.

[0029] Turning to FIG. 2, an exemplary method for data pruning and restoration is indicated generally by the reference numeral 200. The method 200 includes a start block 205 that passes control to a function block 210. The function block 210 receives a picture, and passes control to a function block 215. The function block 215 performs pre-processing of the picture to remove, in the spatial domain, portions (e.g., a line, curve, region, etc.) from the picture, and passes control to a function block 220. The function block 220 codes the picture with the portions removed, and passes control to a function block 225. The function block 225 provides (e.g., transmits) the coded picture to a decoder (or a device having a decoder), and passes control to a function block 230. The function block 230 decodes the coded picture with the portions removed, and passes control to a function block 225. The function block 235 receives the decoded picture with the portions removed, performs post-processing to restore, in the spatial domain, the portions to the decoded picture, and passes control to an end block 299.

[0030] Turning to FIG. 3, an exemplary video encoder to which the present principles may be applied is indicated generally by the reference numeral 300.

[0031] The video encoder 300 includes a frame ordering buffer 310 having an output in signal communication with a non-inverting input of a combiner 385. An output of the combiner 385 is connected in signal communication with a first input of a transformer and quantizer 325. An output of the transformer and quantizer 325 is connected in signal communication with a first input of an entropy coder 345 and a first input of an inverse transformer and inverse quantizer 350. An output of the entropy coder 345 is connected in signal communication with a first non-inverting input of a combiner 390. An output of the combiner 390 is connected in signal communication with a first input of an output buffer 335.

[0032] A first output of an encoder controller 305 is connected in signal communication with a second input of the frame ordering buffer 310, a second input of the inverse transformer and inverse quantizer 350, an input of a picture-type decision module 315, an input of a macroblock-type (MB-type) decision module 320, a second input of an intra prediction module 360, a second input of a deblocking filter 365, a first input of a motion compensator 370, a first input of a motion estimator 375, and a second input of a reference picture buffer 380.

[0033] A second output of the encoder controller 305 is connected in signal communication with a first input of a Supplemental Enhancement Information (SEI) inserter 330, a second input of the transformer and quantizer 325, a second input of the entropy coder 345, a second input of the output buffer 335, and an input of the Sequence Parameter Set (SPS) and Picture Parameter Set (PPS) inserter 340.

[0034] A first output of the picture-type decision module 315 is connected in signal communication with a third input of a frame ordering buffer 310. A second output of the picture-type decision module 315 is connected in signal communication with a second input of a macroblock-type decision module 320.

[0035] An output of the Sequence Parameter Set (SPS) and Picture Parameter Set (PPS) inserter 340 is connected in signal communication with a third non-inverting input of the combiner 390.

[0036] An output of the inverse quantizer and inverse transformer 350 is connected in signal communication with a first non-inverting input of a combiner 319. An output of the combiner 319 is connected in signal communication with a first input of the intra prediction module 360 and a first input of the deblocking filter 365. An output of the deblocking filter 365 is connected in signal communication with a first input of a reference picture buffer 380. An output of the reference picture buffer 380 is connected in signal communication with a second input of the motion estimator 375. A first output of the motion estimator 375 is connected in signal communication with a second input of the motion compensator 370. A second output of the motion estimator 375 is connected in signal communication with a third input of the entropy coder 345.

[0037] An output of the motion compensator 370 is connected in signal communication with a first input of a switch 397. An output of the intra prediction module 360 is connected in signal communication with a second input of the switch 397. An output of the macroblock-type decision module 320 is connected in signal communication with a third input of the switch 397. The third input of the switch 397 determines whether or not the "data" input of the switch (as compared to the control input, i.e., the third input) is to be provided by the motion compensator 370 or the intra prediction module 360. The output of the switch 397 is connected in signal communication with a second non-inverting input of the combiner 319 and with an inverting input of the combiner 385.

[0038] Inputs of the frame ordering buffer 310 and the encoder controller 305 are available as input of the encoder 300, for receiving an input picture 301. Moreover, an input of the Supplemental Enhancement Information (SEI) inserter 330 is available as an input of the encoder 300, for receiving metadata. An output of the output buffer 335 is available as an output of the encoder 300, for outputting a bitstream.

[0039] Turning to FIG. 4, an exemplary video decoder to which the present principles may be applied is indicated generally by the reference numeral 400.

[0040] The video decoder 400 includes an input buffer 410 having an output connected in signal communication with a first input of the entropy decoder 445. A first output of the entropy decoder 445 is connected in signal communication with a first input of an inverse transformer and inverse quantizer 450. An output of the inverse transformer and inverse quantizer 450 is connected in signal communication with a second non-inverting input of a combiner 425. An output of the combiner 425 is connected in signal communication with a second input of a deblocking filter 465 and a first input of an intra prediction module 460. A second output of the deblocking filter 465 is connected in signal communication with a first input of a reference picture buffer 480. An output of the reference picture buffer 480 is connected in signal communication with a second input of a motion compensator 470.

[0041] A second output of the entropy decoder 445 is connected in signal communication with a third input of the motion compensator 470 and a first input of the deblocking filter 465. A third output of the entropy decoder 445 is connected in signal communication with an input of a decoder controller 405. A first output of the decoder controller 405 is connected in signal communication with a second input of the entropy decoder 445. A second output of the decoder controller 405 is connected in signal communication with a second input of the inverse transformer and inverse quantizer 450. A third output of the decoder controller 405 is connected in signal communication with a third input of the deblocking filter 465. A fourth output of the decoder controller 405 is connected in signal communication with a second input of the intra prediction module 460, with a first input of the motion compensator 470, and with a second input of the reference picture buffer 480.

[0042] An output of the motion compensator 470 is connected in signal communication with a first input of a switch 497. An output of the intra prediction module 460 is connected in signal communication with a second input of the switch 497. An output of the switch 497 is connected in signal communication with a first non-inverting input of the combiner 425.
An input of the input buffer 410 is available as an input of the decoder 400, for receiving an input bitstream. A first output of the deblocking filter 465 is available as an output of the decoder 400, for outputting an output picture.

[0043] In accordance with the present principles, we propose a new scheme for data pruning to reduce the compressed video bit rate while, at the same time, preserving a much higher quality of the decoded and reconstructed picture at the decoder end. In further detail with respect to an embodiment, the original images/pictures are pruned to a smaller size prior to compression and then interpolated to their original size after decoding. This interpolation uses information about which lines, segments, or regions were pruned during the data pruning step to return the video image/pictures to their original size. Pruning is performed prior to encoding by dropping lines, segments, and/or regions in the spatial domain, according to one of several possible evaluation schemes and then restoring these lines, segments, and/or regions after decoding the bitstream back into the spatial domain. In an embodiment, edge-directed interpolation is utilized to reconstruct the decoded picture to its original size. Since we avoid filtering the remaining pixels in the image/video pictures, the reconstructed pictures can achieve higher quality from a lower bit-rate.

[0044] In one embodiment, data pruning is performed in a manner in which only rows or columns are dropped, so only up-sampling in one direction is needed in these areas. Although described in such a context, it would be apparent to those skilled in this and related arts that other means for dropping sections in the spatial domain may also be used prior to encoding, while maintaining the present principles. Thus, the lines dropped may be rows or columns, but are not limited to only the preceding. For example, the lines dropped may be diagonal or may be non-linear curves instead of lines and/or in addition to lines. Furthermore, the same principles would apply and are within the scope of the present principles in that the requisite unit to be pruned is not a spatial domain line or curve, but rather a section or region. Thus, the description of the present principles provided herein in the context of lines should not so limit the scope of the invention to lines only, but to any selection of lines, curves, sections, and/or regions within the spatial domain to be dropped prior to encoding (and consequently restored after decoding). Thus, given the teachings of the present principles provided herein, these and other variations of the present principles are contemplated thereby and are readily capable of being implemented by one of ordinary skill in this and related arts, while maintaining the present principles. It is to be appreciated that as used herein, the term "region" may represent one or more of a line, curve, section, and/or region.

[0045] For illustrative purposes, we focus on how to adjust New Edge-Directed Interpolation (NEDI, specifically NEDI-4) for interpolation. Turning to FIG. 5A, model parameters of high-order edge-directed interpolation relating to NEDI-4 and NEDI-8 are indicated generally by the reference numeral 500. As shown therein, two passes are performed using the quincunx sub-lattice. In the first pass, the high resolution pixels at the cross of any four nearest low resolution pixels are interpolated from their four nearest pixels by using an adaptive fourth-order interpolator. The remaining high resolution pixels are interpolated with the same algorithm in the second pass, using the low resolution pixels and the pixels interpolated in the first pass. The high resolution model parameters of the interpolator are estimated from the lower resolution model parameters.

Data Pruning



[0046] Data pruning is implemented by dropping lines in the pictures. A particular case would be that the line is a straight line. In that case, a line can be either a row or a column, or any straight line with direction (as seen in FIG. 6). Turning to FIG. 6, an example of data pruning by dropping arbitrary lines is indicated generally by the reference numeral 600. The data pruning 600 involves a frame 610, a pruned line 620, and a pruned curve 630.

[0047] Turning to FIG. 7, an example of data pruning by dropping columns and rows in one frame is indicated generally by the reference numeral 700. The data pruning 700 involves a frame 710, a dropped column 720, and a dropped row 730.

[0048] Turning to FIG. 8, an example of data pruning by dropping the same columns and rows for a whole group of pictures (GOP) and/or scene is indicated generally by the reference numeral 800. The data pruning 800 involves a first frame 810, middle frames 820, and a last frame 830, dropped rows 840, and dropped columns 850.

[0049] Turning to FIG. 9A, lines for indicating the dropped columns and rows in the first frame of the well-known Akiyo sequence are indicated generally by the reference numeral 900. Turning to FIG. 9B, the pruned frame corresponding to the example of FIG. 9A is indicated generally by the reference numeral 950.

[0050] We need to know the location of the dropped line at the decoder, because the decoder needs to interpolate the missing pixels. There is a tradeoff between signaling the location of the pixels of the dropped line and the overhead for keeping track of this dropped line. In one embodiment, we drop only columns and rows in order to significantly reduce the overhead to indicate the line. In this case, only 1 bit per row/column is used to keep track, thus indicating whether the line is dropped or not (as seen in FIG. 7 for dropping the columns and rows in one frame, in FIG. 8 for dropping the columns and rows in one GOP/scene, in FIG. 9A with white lines indicating the dropped columns and rows, and in FIG. 9B for the pruned frame). In another embodiment, we drop only odd lines and keep even lines untouched. In one embodiment, all the pictures in one GOP/scene have the same dropping patterns in order to further reduce the overhead. To decide which line is to be dropped first, in one embodiment, we assume no compression is involved and we use a distortion criterion. We first set up a final target. The target can be distortion, percentage of dropping pixels and/or any complexity constraint. Of course, given the teachings of the present principles provided herein, the target is not limited to the preceding and, thus, other targets may also be used in accordance with the present principles, while maintaining the present principles. In one embodiment, presuming one line is dropped, we interpolate the dropped line without considering compression. Then, we compare the MSE (Mean Square Error) of the interpolated line to the same line in the original image. The line that has the minimum MSE is selected to be dropped first. In another embodiment, we take both the distortion and the bit-rate of the line into consideration. In one embodiment, the line bit-rate can be approximated using its variance.

Optimal Data Pruning Embodiment



[0051] Turning to FIG. 10, an exemplary system for data pruning based compression is indicated generally by the reference numeral 1000. The system 1000 includes a data pruner 1010, a data store or compressor 1020, and an interpolator 1030. While shown as separate elements, one or more of the data pruner 1010, data store or compressor 1020, or interpolator 1030 may be combined into an integrated element. For example, in an embodiment, the data pruner 1010 may be included within the data store or compressor 1020. The data store or compressor 1020 may be considered to be an encoder for the purposes of the present principles. The original frame I of size M × N is pruned to frame P of smaller size (M - Mp)×(N - Np), where Mp and Np are the number of dropped rows and columns, respectively. The purpose of data pruning is to reduce the number of bits to represent the stored or compressed frame P'. P' is then interpolated to I' to reconstruct to the original frame size. The compression stage considered here is the MPEG-4 AVC Standard codec, with the encoder as shown in FIG. 3 and the decoder as shown in FIG. 4. However, as noted above, the present principles are not limited to solely the MPEG-4 AVC Standard and, thus, may be applied to other video coding standards, recommendations, and extensions thereof, while maintaining the present principles.

[0052] Only the even rows and columns are considered to be dropped while the odd rows and columns are kept for later interpolation. Turning to FIG. 11, an exemplary system for a data pruning phase is indicated generally by the reference numeral 1100. The system 1100 includes a decimator 1110 and an interpolator 1120. In the data pruning phase, the original frame I is decimated by the decimator 1110 to the low resolution (LR) frame Il by a ratio of 1 × 2 , 2 × 1 and 2 × 2 for the respective cases of dropping only columns, dropping only rows, and dropping both columns and rows. Then, frame Il, is interpolated by the interpolator 1120 to the high resolution (HR) frame Ih. Due to the interpolation, the reconstructed frame is different than its original frame. The rows and columns in Ih which have the least error comparing to their corresponding rows and columns in I are chosen to be dropped. The mean squared error (MSE) is defined as the mean square error between Ih and I.

[0053] Given the target MSEmax, the data pruning is optimized to drop the most pixels while still keeping the overall MSE less than MSEmax. The location of the dropped rows and columns is respectively indicated by αm and αn. The ith even line (row or column) is dropped if the value of αi is 1; otherwise it will be kept. These indicators are stored as side information in the coded bit-stream. The mean square line error (MSLE) for one dropped line is defined as the mean square error between Ih and I for pixels on that line only. Lines with a smaller MSLE have a higher priority to be dropped than lines with a higher MSLE . Presume that the Mp rows and Np columns with the smallest MSLE are dropped and the maximum MSLE of these lines is MSLEmax. The overall MSE becomes the averaged MSE of all dropped pixels. An upper limit of MSE is achieved based on, for example, the number of dropped rows Mp, dropped columns Np, and MSLEmax. This upper limit should be less than MSEmax as the condition to find the optimal data pruning scheme.

[0054] Turning to FIG. 12, an exemplary method for optimal data pruning is indicated generally by the reference numeral 1200. The method 1200 includes a start block 1201 that passes control to a function block 1203. The function block 1203 inputs a frame or GOP, and passes control to a function block 1206, a function block 1209, and a function block 1212. The function block 1206 removes the even rows, and passes control to a function block 1015. The function block 1215 performs interpolation based on, for example, NEDI-6, and passes control to a function block 1224. The function block 1224 finds the error between the original and reconstructed rows (1), and passes control to a function block 1233.

[0055] The function block 1209 removes the even rows, and passes control to a function block 1218. The function block 1218 performs interpolation based on, for example, NEDI-6, and passes control to a function block 1227. The function block 1227 finds the error between the original and reconstructed columns (2), and passes control to the function block 1233.

[0056] The function block 1212 removes even rows and columns, and passes control to a function block 1221. The function block 1221 performs interpolation based on, for example, NEDI-8, and passes control to a function block 1230. The function block 1230 finds the error between the original and reconstructed rows/columns (3), and passes control to the function block 1233.

[0057] The function block 1233 sets the number (no.) of dropped rows M_d = 0, sets the number of dropped columns N_d = 0, sets the maximum dropped pixels Pel = 0, and passes control to a decision block 1236. The decision block 1236 determines whether or not M_d = 0. If so, then control is passed to a function block 1242. Otherwise, control is passed to a decision block 1239.

[0058] The function block 1242 chooses to drop the N_d columns with the least error in (1), i.e., the error calculated per function block 1224, and passes control to a function block 1246. The function block 1246 calculates the error MSE and number of dropped pixels Pel, and passes control to a decision block 1252. The decision block 1252 determines whether or not MSE<MSE_max and Pel>Pel_max. If so, then control is passed to a function block 1255. Otherwise, control is passed to a function block 1270.

[0059] The function block 1255 sets Pel_max = Pel, sets M_d_opt = M_d, sets N_d_opt = N_d, and passes control to a function block 1258. The function block 1258 determines whether or not N_d < N. If so, then control is passed to a function block 1264. Otherwise, control is passed to a function block 1261.

[0060] The function block 1264 sets N_d = N_d + 16, and returns control to the decision block 1236.

[0061] The function block 1261 determines whether or not M_d < M. If so, then control is passed to a function block 1267. Otherwise, control is passed to a function block 1270.

[0062] The decision block 1239 determines whether or not N_d = 0. If so, then control is passed to a function block 1245. Otherwise, control is passed to a function block 1249.

[0063] The function block 1245 chooses to drop the M_d columns with the least error in (2), i.e., the error calculated per function block 1227, and passes control to the function block 1246.

[0064] The function block 1249 chooses to drop the M_d columns and N_d rows with the least error in (3), i.e., the error calculated per function block 1230, and passes control to the function block 1046.

[0065] The function block 1270 outputs M_d_opt and N_d_opt, and passes control to an end block 1299.

[0066] Turning to FIG. 13, an exemplary method for optimal data restoring is indicated generally by the reference numeral 1300. The method 1300 includes a start block 1305 that passes control to a function block 1310. The function block 1310 inputs a pruned frame and/or group of pictures (GOP), and a list of dropped frames, and passes control to a function block 1315. The function block 1315 resizes the frames to their original size, and passes control to a loop limit block 1320. The loop limit block 1320 loops over the pruned data, and passes control to a decision block 1325. The decision block 1325 determines whether or not an even row and column has been removed. If so, then control is passed to a function block 1330. Otherwise, control is passed to a decision block 1340.

[0067] The function block 1330 interpolates the row and column with NEDI-8, and passes control to a loop limit block 1335.

[0068] The loop limit block 1335 ends the loop, and passes control to a function block 1360. The function block 1360 outputs a reconstructed frame/GOP, and passes control to an end block 1399.

[0069] The decision block 1340 determines whether or not an even row has been removed. If so, then control is passed to a function block 1345. Otherwise, control is passed to a decision block 1350.

[0070] The function block 1345 interpolates the row with NEDI-6, and passes control to the loop limit block 1335.

[0071] The decision block 1350 determines whether or not an even column has been removed. If so, then control is passed to a function block 1355. Otherwise, control is passed to the loop limit block 1335.

[0072] The function block 1355 interpolates the column with NEDI-6, and passes control to the loop limit block 1335.

High Order Edge-Directed Interpolation



[0073] By interpolation we mean any method that is able to estimate/fill missing pixels from the surrounding pixels of the same or temporally related frames. Such methods can be classical interpolation, like the bi-cubic method, newer methods like NEDI, or other kind of methods like inpainting, which fills holes of missing pixels from their surroundings. Of course, interpolation in accordance with the present principles is not limited to only the preceding described interpolation techniques and, thus, other interpolation techniques may be also be used, while maintaining the present principles.

[0074] For interpolation at a decoder, we propose to apply a high-order edge-directed interpolation so more directions can be adapted. Our method is different from that described in a prior art approach relating to covariance of the low resolution image, which uses only 4 neighboring pixels and can only be applied for an upsizing ratio of 2 × 2. We adapt it to be used for the proposed data pruning scheme. In an embodiment, the sixth-order and eighth- order edge-directed interpolation are used for the cases of dropping lines in only one dimension (only rows or only columns) and 2 dimensions (both rows and columns), respectively.

Embodiment: NEDI-6



[0075] The algorithm for dropping only columns is discussed because the same algorithm can be applied for the cases of dropping only columns and dropping only rows. First, the columns of P' corresponding to the odd columns in the original frame I are extracted to form the low resolution frame P'l of size

Columns of P'l are mapped to the odd columns of the HR frame P'h of size M × N. The even columns of P'h are interpolated from the odd columns by a sixth-order interpolation as explained herein.

[0076] The interpolated pixel is the summation of the weighted 6-neighbor surrounding pixels as shown in FIG. 5B. Turning to FIG. 5B, model parameters of high-order edge-directed interpolation relating to NEDI-6 are indicated generally by the reference numeral 500. These weights are the model parameters of the high resolution frame. The optimal model parameters are obtained by minimizing the MSE between the interpolated pixels and the original pixels with the assumption that the model parameters are constant in a local window. Applying the geometric duality assumption, the model parameters are considered constant for different scales and can be estimated from the low resolution pixels. As shown in FIG. 5B, the model parameters of the low resolution frame, with the availability of the original pixels and interpolated pixels, can be obtained by minimizing the error between them. These low resolution model parameters are then used as the high resolution model parameters to interpolate the missing high resolution pixels. Finally, the column indicator is used to choose the pixels from the compressed pruned frame or interpolated frame to form the final reconstructed frame. If the column indicator value is 1, then the reconstructed column is taken from the interpolated frame. Otherwise, if the column indicator value is 0, then the reconstructed column is taken from the compressed pruned frame.

Embodiment: NEDI-8



[0077] Similar to NEDI-6, the pixels in P' corresponding to the low resolution pixels of the down-sampling ratio 2 × 2 in the original frame I are extracted to form the LR frame P'l of size

The interpolation is performed using NEDI-4 for the first round and NEDI-8 for the second round. With the availability of all 8-neighboring pixels in the second round, NEDI-8 is implemented to get extra information from 4 directions. The advantage of this round is that the model parameters can be directly estimated from its high resolution pixels, so the over-fitting problem of NEDI-4 is lessened. NEDI-8 is applied to all the missing high resolution pixels in order to be consistent. The fourth-order model parameters h4 and eighth-order model parameters h8 for HR scale are shown in FIG. 5A. The optimal h8 is calculated by a similar algorithm to the one employed for NEDI-6.

[0078] A description will now be given of some of the many attendant advantages/features of the present invention, some of which have been mentioned above. For example, one advantage/feature is an apparatus having a data pruner for pre-processing a picture prior to, and in preparation for, compression by encoding. The data pruner selectively removes, in the spatial domain, portions within the picture.

[0079] Another advantage/feature is the apparatus having the data pruner as described above, wherein the data pruner is included in an encoder.

[0080] Still another advantage/feature is the apparatus having the data pruner as described above, further comprising an encoder coupled to the data pruner for encoding the picture with the portions removed.

[0081] Also another advantage/feature is the apparatus having the data pruner as described above, wherein side information indicating a location of the portions within the picture is one of explicitly or implicitly transmitted to a decoder or other device.

[0082] Yet another advantage/feature is the apparatus having the data pruner as described above, wherein the portions includes at least one of a straight line, a non-linear curve, and a region.

[0083] Moreover, another advantage/feature is the apparatus having the data pruner as described above, wherein only even or odd lines are removed by the data pruner.

[0084] Further, another advantage/feature is the apparatus having the data pruner as described above, wherein the picture is included in a group of pictures, and all pictures within the group of pictures are assigned a same removal pattern.

[0085] Also, another advantage/feature is the apparatus having the data pruner as described above, wherein the portions selectively removed is determined responsive to an evaluation of at least one of an image distortion metric, a rate distortion metric, a dropped pixel metric, and a complexity metric.

[0086] Additionally, another advantage/feature is the apparatus having the data pruner as described above, wherein 6-order interpolation is used for removing the portions in one dimension, and 8-order interpolation is used for removing the portions in two dimensions.

[0087] These and other features and advantages of the present principles may be readily ascertained by one of ordinary skill in the pertinent art based on the teachings herein. It is to be understood that the teachings of the present principles may be implemented in various forms of hardware, software, firmware, special purpose processors, or combinations thereof.

[0088] Most preferably, the teachings of the present principles are implemented as a combination of hardware and software. Moreover, the software may be implemented as an application program tangibly embodied on a program storage unit. The application program may be uploaded to, and executed by, a machine comprising any suitable architecture. Preferably, the machine is implemented on a computer platform having hardware such as one or more central processing units ("CPU"), a random access memory ("RAM"), and input/output ("I/O") interfaces. The computer platform may also include an operating system and microinstruction code. The various processes and functions described herein may be either part of the microinstruction code or part of the application program, or any combination thereof, which may be executed by a CPU. In addition, various other peripheral units may be connected to the computer platform such as an additional data storage unit and a printing unit.

[0089] It is to be further understood that, because some of the constituent system components and methods depicted in the accompanying drawings are preferably implemented in software, the actual connections between the system components or the process function blocks may differ depending upon the manner in which the present principles are programmed. Given the teachings herein, one of ordinary skill in the pertinent art will be able to contemplate these and similar implementations or configurations of the present principles.

[0090] Although the illustrative embodiments have been described herein with reference to the accompanying drawings, it is to be understood that the present principles is not limited to those precise embodiments, and that various changes and modifications may be effected therein by one of ordinary skill in the pertinent art without departing from the scope of the present principles. All such changes and modifications are intended to be included within the scope of the present principles as set forth in the appended claims.


Claims

1. A method, comprising:
pre-processing (215) a picture prior to, and in preparation for, compression by encoding, wherein said pre-processing step comprises selectively removing, based on location of pixels within the spatial domain of the picture, and not in the frequency domain, portions of said picture and leaving the remaining portions of said picture to be encoded using a data pruner having a processor (1200), characterized in transmitting side information indicating a location of the removed portions within the picture to a decoder or other device, wherein the selectively removed portions are diagonal lines of said picture, and wherein the diagonal lines are neither rows nor columns and encoding the pre-processed picture.
 
2. An apparatus for implementing the method of claim 1, comprising:
a data pruner (110, 1010) for pre-processing a picture prior to, and in preparation for, compression by encoding, wherein said data pruner selectively removes based on location of pixels, within the spatial domain of the picture and not in the frequency domain, portions of said picture and leaving the remaining portions of said picture to be encoded, characterized in that side information indicating a location of the portions within the picture transmitted to a decoder or other device and wherein the data pruner (110, 1010) selectively removes portions which are diagonal lines of said picture, and wherein the diagonal lines are neither rows nor columns, and an encoder.
 
3. The apparatus of claim 2, wherein the data pruner is comprised in the encoder.
 
4. The apparatus of claim 2, wherein the encoder is coupled to said data pruner for encoding the picture with the portions removed.
 
5. The apparatus of claims 2, 3, 4, wherein the portions selectively removed is determined responsive to an evaluation of at least one of an image distortion metric, a rate distortion metric, a dropped pixel metric, and a complexity metric (1246, 1252).
 
6. A method, comprising:

decompressing a received picture,

receiving (235) the decompressed picture subsequent to decompression by decoding; and

post-processing (235), using a data restorer having a processor, the decompressed picture by selectively restoring, based on location of pixels within the spatial domain of the picture, and not on the frequency domain, portions of said picture, from the remaining portions of said picture that was encoded, based on information indicating a removal of the portions prior to a previously performed encoding process (1300), characterized in receiving side information indicating a location of the removed portions within the picture from a decoder or other device and wherein the selectively restored portions are diagonal lines within said decompressed picture, wherein the diagonal lines are neither rows nor columns.


 
7. An apparatus for implementing the method of claim 6, comprising:

a decoder for decompressing a received picture and

a data restorer (140) for receiving a decompressed picture subsequent to decompression by decoding based on location of pixels, and post-processing the decompressed picture by selectively restoring, in the spatial domain of the picture and not in the frequency domain, portions of said picture from the remaining portions of said picture that was encoded based on information indicating a removal of the portions prior to a previously performed encoding process, characterized in that side information indicating a location of the removed portions within the picture is received from a decoder or other device and wherein the data restorer (140) selectively restores the portions which are diagonal lines within said decompressed picture, and wherein the diagonal lines are neither rows nor columns.


 
8. The apparatus of claim 7, wherein said data restorer is comprised in the decoder.
 
9. The apparatus of claim 7, wherein the decoder is coupled to said data restorer for decoding the picture with the portions removed.
 
10. The apparatus of claims 2 to 5 or 7 to 9, wherein the picture is comprised in a group of pictures, and all pictures within the group of pictures are assigned a same removal pattern (1310).
 
11. The apparatus of claims 2 to 5 or 7 to 9, wherein 6-order interpolation is used for removing or restoring the portions in one dimension, and 8-order interpolation is used for removing or restoring the portions in two dimensions (1345, 1330).
 
12. A storage media having video signal data encoded thereupon, comprising:
an encoded picture pre-processed prior to compression by encoding, wherein the picture is pre-processed to selectively remove, based on location of pixels within the spatial domain of the picture, and not on the frequency domain, portions of said picture and leaving the remaining portion of said picture to be encoded, characterized in that side information indicating a location of the removed portions is stored, and the picture is pre-preprocessed to selectively remove the portions which are diagonal lines within the picture, wherein the diagonal lines are neither rows nor columns.
 


Ansprüche

1. Verfahren, das umfasst:
Vorverarbeiten (215) eines Bilds vor der und in Vorbereitung auf die Kompression durch Codieren, wobei der Schritt des Vorverarbeitens das selektive Entfernen von Abschnitten des Bilds auf der Grundlage des Orts von Pixeln in dem räumlichen Bereich des Bilds und nicht in dem Frequenzbereich und das Belassen der verbleibenden Abschnitte des zu codierenden Bilds unter Verwendung einer Datenlöscheinrichtung mit einem Prozessor (1200) umfasst, gekennzeichnet durch das Senden von Nebeninformationen, die einen Ort der entfernen Abschnitte in dem Bild angeben, an einen Decodierer oder an eine andere Vorrichtung, wobei die selektiv entfernten Abschnitte Diagonalen des Bilds sind und wobei die Diagonalen weder Zeilen noch Spalten sind und das vorverarbeitete Bild codieren.
 
2. Vorrichtung zum Implementieren des Verfahrens nach Anspruch 1, wobei die Vorrichtung umfasst:
eine Datenlöscheinrichtung (110, 1010) zum Vorverarbeiten eines Bilds vor der und in Vorbereitung auf die Kompression durch Codieren, wobei die Datenlöscheinrichtung Abschnitte des Bilds auf der Grundlage des Orts von Pixeln in dem räumlichen Bereich des Bilds und nicht in dem Frequenzbereich selektiv entfernt und die verbleibenden Abschnitte des zu codierenden Bilds belässt, dadurch gekennzeichnet, dass Nebeninformationen, die einen Ort der Abschnitte in dem Bild angeben, an einen Decodierer oder an eine andere Vorrichtung gesendet werden und dass die Datenlöscheinrichtung (110, 1010) Abschnitte, die Diagonalen des Bilds sind, selektiv entfernt und dass die Diagonalen weder Zeilen noch Spalten sind, und einen Codierer
 
3. Vorrichtung nach Anspruch 2, wobei die Datenlöscheinrichtung in dem Codierer enthalten ist.
 
4. Vorrichtung nach Anspruch 2, wobei der Codierer zum Codieren des Bilds mit den entfernten Abschnitten mit der Datenlöscheinrichtung gekoppelt ist.
 
5. Vorrichtung nach Anspruch 2, 3, 4, wobei die selektiv entfernten Abschnitte als Reaktion auf eine Bewertung einer Bildverzerrungsmetrik und/oder einer Ratenverzerrungsmetrik und/oder einer Metrik verworfener Pixel und/oder einer Komplexitätsmetrik bestimmt werden (1246, 1252).
 
6. Verfahren, das umfasst:

Dekomprimieren eines empfangenen Bilds,

Empfangen (235) des dekomprimierten Bilds nach der Dekompression durch Decodieren; und

Nachverarbeiten (235) des dekomprimierten Bilds durch selektives Wiederherstellen von Abschnitten des Bilds auf der Grundlage des Orts von Pixeln in dem räumlichen Bereich des Bilds und nicht in dem Frequenzbereich aus den verbleibenden Abschnitten des Bilds, das codiert war, auf der Grundlage von Informationen, die eine Entfernung der Abschnitte vor einem zuvor ausgeführten Codierungsprozess angeben, unter Verwendung einer Datenwiederherstellungseinrichtung mit einem Prozessor (1300), gekennzeichnet durch das Empfangen von Nebeninformationen, die einen Ort der entfernen Abschnitte in dem Bild angeben, von einen Decodierer oder von einer anderen Vorrichtung, und wobei die selektiv wiederhergestellten Abschnitte Diagonalen in dem dekomprimierten Bild sind, wobei die Diagonalen weder Zeilen noch Spalten sind.


 
7. Vorrichtung zum Implementieren des Verfahrens nach Anspruch 6, wobei die Vorrichtung umfasst:

einen Decodierer zum Dekomprimieren eines empfangenen Bilds und

eine Datenwiederherstellungseinrichtung (140) zum Empfangen eines dekomprimierten Bilds nach der Dekompression durch Decodieren auf der Grundlage des Orts von Pixeln und zum Nachverarbeiten des dekomprimierten Bilds durch selektives Wiederherstellen von Abschnitten des Bilds in dem räumlichen Bereich des Bilds und nicht in dem Frequenzbereich aus den verbleibenden Abschnitten des Bilds, das codiert war, auf der Grundlage von Informationen, die eine Entfernung der Abschnitte vor einem zuvor ausgeführten Codierungsprozess angeben, dadurch gekennzeichnet, dass von einem Decodierer oder von einer anderen Vorrichtung Nebeninformationen empfangen werden, die einen Ort der entfernen Abschnitte in dem Bild angeben, und dass die Datenwiederherstellungseinrichtung (140) die Abschnitte, die Diagonalen in dem dekomprimierten Bild sind, selektiv wiederherstellt und dass die Diagonalen weder Zeilen noch Spalten sind.


 
8. Vorrichtung nach Anspruch 7, wobei die Datenwiederherstellungseinrichtung in dem Decodierer enthalten ist.
 
9. Vorrichtung nach Anspruch 7, wobei der Decodierer zum Decodieren des Bilds mit den entfernten Abschnitten mit der Datenwiederherstellungseinrichtung gekoppelt ist.
 
10. Vorrichtung nach Anspruch 2 bis 5 oder 7 bis 9, wobei das Bild in einer Bildgruppe enthalten ist und wobei allen Bildern in der Bildgruppe ein selbes Entfernungsmuster (1310) zugewiesen ist.
 
11. Vorrichtung nach Anspruch 2 bis 5 oder 7 bis 9, wobei zum Entfernen oder Wiederherstellen der Abschnitte in einer Dimension eine Interpolation 6. Ordnung verwendet wird und zum Entfernen oder Wiederherstellung der Abschnitte in zwei Dimensionen eine Interpolation 8. Ordnung verwendet wird (1345, 1330).
 
12. Speichermedien mit darin codierten Videosignaldaten, wobei die Speichermedien umfassen:
ein codierten Bild, das vor der Kompression durch Codierung vorverarbeitet worden ist, wobei das Bild vorverarbeitet wird, um auf der Grundlage des Orts von Pixeln in dem räumlichen Bereich des Bilds und nicht in dem Frequenzbereich Abschnitte des Bilds selektiv zu entfernen und die verbleibenden Abschnitte des zu codierenden Bilds zu belassen, dadurch gekennzeichnet, dass Nebeninformationen, die einen Ort der entfernen Abschnitte angeben, gespeichert sind und dass das Bild vorverarbeitet wird, um die Abschnitte, die Diagonalen in dem Bild sind, selektiv zu entfernen, wobei die Diagonalen weder Zeilen noch Spalten sind.
 


Revendications

1. Procédé comprenant :
le prétraitement (215) d'une image préalablement à, et en préparation de, la compression par codage, dans lequel ladite étape de prétraitement comprend la suppression sélective, selon l'emplacement de pixels dans le domaine spatial de l'image, et non dans le domaine de fréquence, de parties de ladite image et la conservation des parties restantes de ladite image à coder à l'aide d'un dispositif de coupe de données doté d'un processeur (1200), caractérisé par la transmission d'informations complémentaires indiquant l'emplacement des parties supprimées dans l'image à un décodeur ou un autre dispositif, dans lequel les parties supprimées de façon sélective sont des lignes diagonales de ladite image, et dans lequel les lignes diagonales ne sont ni des rangées ni des colonnes, et le codage de l'image prétraitée.
 
2. Appareil pour la mise en oeuvre du procédé selon la revendication 1, comprenant :
un dispositif de coupe de données (110, 1010) pour le prétraitement d'une image préalablement à, et en préparation de, la compression par codage, dans lequel ledit dispositif de coupe de données supprime de façon sélective, selon l'emplacement de pixels, dans le domaine spatial de l'image et non dans le domaine de fréquence, des parties de ladite image et conserve les parties restantes de ladite image à coder, caractérisé en ce que des informations complémentaires indiquant l'emplacement des parties dans l'image sont transmises à un décodeur ou un autre dispositif, et dans lequel le dispositif de coupe d'images (110, 1010) supprime de façon sélective des parties qui sont des lignes diagonales de ladite image, et dans lequel les lignes diagonales ne sont ni des rangées ni des colonnes, et un encodeur.
 
3. Appareil selon la revendication 2, dans lequel le dispositif de coupe de données est compris dans l'encodeur.
 
4. Appareil selon la revendication 2, dans lequel l'encodeur est couplé audit dispositif de coupe de données pour le codage de l'image avec les parties supprimées.
 
5. Appareil selon les revendications 2, 3, 4, dans lequel les parties supprimées de façon sélective sont déterminées en réponse à une évaluation d'une métrique de distorsion d'image et / ou d'une métrique de distorsion de taux et / ou d'une métrique de suppression de pixels et / ou d'une métrique de complexité (1246, 1252).
 
6. Procédé comprenant :
la décompression d'une image reçue,
la réception (235) de l'image décompressée après la décompression par décodage ; et
le post-traitement (235), à l'aide d'un dispositif de restauration de données doté d'un processeur, de l'image décompressée par la restauration sélective, selon l'emplacement de pixels dans le domaine spatial de l'image, et non dans le domaine de fréquence, de parties de ladite image, à partir des parties restantes de l'image qui a été codée, d'après des informations indiquant une suppression des parties avant un processus de codage préalablement effectué (1300), caractérisé par la réception d'informations complémentaires indiquant l'emplacement des parties supprimées dans l'image à partir d'un décodeur ou d'un autre dispositif et dans lequel les parties restaurées de façon sélective sont des lignes diagonales dans ladite image décompressée, dans lequel les lignes diagonales ne sont ni des rangées ni des colonnes.
 
7. Appareil pour la mise en œuvre du procédé selon la revendication 6, comprenant :

un décodeur pour décompresser une image reçue et

un dispositif de restauration de données (140) pour recevoir une image décompressée après la décompression par décodage selon l'emplacement de pixels, et le post-traitement de l'image décompressée par la restauration sélective, dans le domaine spatial de l'image et non dans le domaine de fréquence, de parties de ladite image à partir des parties restantes de ladite image qui a été codée d'après des informations indiquant une suppression des parties avant un processus de codage préalablement effectué, caractérisé en ce que des informations complémentaires indiquant l'emplacement des parties supprimées dans l'image sont reçues à partir d'un décodeur ou d'un autre dispositif et dans lequel le dispositif de restauration de données (140) restaure de façon sélective les parties qui sont des lignes diagonales dans ladite image décompressée, et dans lequel les lignes diagonales ne sont ni des rangées ni des colonnes.


 
8. Appareil selon la revendication 7, dans lequel ledit dispositif de restauration de données est compris dans le décodeur.
 
9. Appareil selon la revendication 7, dans lequel le décodeur est couplé audit dispositif de restauration de données pour le décodage de l'image avec les parties supprimées.
 
10. Appareil selon les revendications 2 à 5 ou 7 à 9, dans lequel l'image est comprise dans un groupe d'images, et un même schéma de suppression (1310) est affecté à toutes les images du groupe d'images.
 
11. Appareil selon les revendications 2 à 5 ou 7 à 9, dans lequel une interpolation d'ordre 6 est utilisée pour supprimer ou restaurer les parties dans une dimension et une interpolation d'ordre 8 est utilisée pour supprimer ou restaurer les parties dans deux dimensions (1345, 1330).
 
12. Support de stockage contenant des données de signal vidéo codées, comprenant :
une image codée prétraitée préalablement à la compression par codage, où l'image est prétraitée pour supprimer de façon sélective, selon l'emplacement de pixels dans le domaine spatial de l'image, et non dans le domaine de fréquence, des parties de ladite image et conserver la partie restante de ladite image à coder, caractérisé en ce que des informations complémentaires indiquant l'emplacement des parties supprimées sont stockées, et l'image est prétraitée pour supprimer de façon sélective les parties qui sont des lignes diagonales dans l'image, où les lignes diagonales ne sont ni des rangées ni des colonnes.
 




Drawing



































Cited references

REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description