CN103533375A - Video compression method and video compression device - Google Patents
Video compression method and video compression device Download PDFInfo
- Publication number
- CN103533375A CN103533375A CN201210276252.8A CN201210276252A CN103533375A CN 103533375 A CN103533375 A CN 103533375A CN 201210276252 A CN201210276252 A CN 201210276252A CN 103533375 A CN103533375 A CN 103533375A
- Authority
- CN
- China
- Prior art keywords
- vector
- motion
- pixels
- block
- current macro
- 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.)
- Granted
Links
- 230000006835 compression Effects 0.000 title claims abstract description 38
- 238000007906 compression Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 18
- 239000013598 vector Substances 0.000 claims abstract description 165
- 238000001514 detection method Methods 0.000 claims description 20
- 239000000284 extract Substances 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 13
- 238000011156 evaluation Methods 0.000 description 4
- 229910003460 diamond Inorganic materials 0.000 description 2
- 239000010432 diamond Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Images
Abstract
A video compression method and apparatus. The video compression method comprises the following steps: extracting a plurality of reference motion vectors of a current macro block, wherein the reference motion vectors are provided by pixel blocks with at least two sizes in a reference macro block of the current macro block; identifying a median motion vector from the reference motion vectors, and setting a search starting point of a similar macro block according to the median motion vector; searching a most similar macro block of the current macro block at the beginning of a reference frame according to the similar macro block searching starting point; and obtaining a motion vector for video compression according to the searched most similar macro block and the current macro block.
Description
Technical field
The present invention has disclosed a kind of video-frequency compression method and video compression device.
Background technology
Video compression comprises an inter prediction (inter prediction) technology.Inter prediction comprises: (for example seek out variation between a present frame (current frame) and a reference frame (reference frame), then from this reference frame, estimate a predictive frame according to motion-vector motion-vector/motion vector).That predictive frame can make difference (residues) data volume between this present frame and this predictive frame is accurately minimum-and the video compression discussed is this difference of compression.
As shown in the above, the effect of video compression depends on the accuracy of predictive frame, and the accuracy of predictive frame is affected by motion-vector estimating techniques.The motion-vector estimation of one pattern of encoding about wish, needs first in reference frame, search similar pattern conventionally.Setting about the search starting point of similar pattern in reference frame may appreciable impact video compression effect.
Summary of the invention
A kind of video-frequency compression method and device.
The video-frequency compression method of realizing according to a kind of execution mode comprises: extract a plurality of with reference to motion-vector of a current macro, with reference to motion-vector, the block of pixels by least two kinds of sizes in the reference macroblock of this current macro is provided for these; From these, with reference to motion-vector, pick out a median motion-vector, and according to this median motion-vector, set a similar macro blocks and search starting point; According to this similar macro blocks, search starting point in similar macro blocks of this current macro of reference frame initial search; And according to this similar macro blocks and this current macro of searching, try to achieve the motion-vector of this current macro for video compression.
A video compression device of realizing according to a kind of execution mode comprises: a memory cell and a mobile detection module.In this cell stores one reference frame and a present frame, a current macro is a plurality of with reference to motion-vector.With reference to motion-vector, the block of pixels by least two kinds of sizes in the reference macroblock of this current macro is provided for these.This mobile detection module from this memory cell, extract this current macro these with reference to motion-vector, and with reference to motion-vector, pick out a median motion-vector from these, to set a similar macro blocks according to this median motion-vector, search starting point.According to this similar macro blocks, search starting point, one similar macro blocks of this this current macro of reference frame initial search that this mobile detection module is stored in this memory cell, and try to achieve motion-vector for video compression according to this similar macro blocks and the current macro that search.
For above-mentioned purpose of the present invention, feature and advantage can be become apparent, special embodiment below, and coordinate appended accompanying drawing, be described in detail as follows.
Accompanying drawing explanation
Fig. 1 is with a kind of execution mode of block diagram illustration video compression device;
Fig. 2 diagram similar macro blocks is searched the definition of starting point (Initial Search Point, ISP);
Needed information when Fig. 3 diagram current macro CurMB mobile detection is set similar macro blocks search starting point ISP;
The limit of Fig. 4 diagram one present frame is divided;
Fig. 5 with in the disclosed image compressing method of flowchart text about the mobile detection of a current macro;
Fig. 6 A, Fig. 6 B, Fig. 6 C diagram " three step searches (three step search) " respectively, " rhombus search (diamond search) " and " hexagonal search (hexagonal search) " technology;
Fig. 7 diagram one median vector estimation design Median-of-4;
Fig. 8 diagram one median vector estimation design Median-of-6; And
A kind of method for searching of Fig. 9 diagram, seeks out similar macro blocks BestMatchedMB for the local Search Area 206 at Fig. 2 according to path indication 902.
[main element symbol description]
102～mobile detection module; 104～memory cell;
106～motion compensation module; 108～predictive frame;
The difference of 110～two interframe;
202～present frame; 204～reference frame;
206～local Search Area;
First macro block of 402～present frame upper left corner;
The upper edge area of 404～present frame;
The left edge area of 406～present frame;
The right edge area of 408～present frame;
The non-edge area of 410～present frame;
702,704～numerical value A ... two median persons in D;
802,804,806,808～numerical value A ... four median persons in F;
810, two median persons in 812～numerical value 802,804,806,808;
The indication of 902～path;
A ... F～numerical value;
Cmp1, cmp2～comparator;
Cycle1 ... Cycle5～the first ... the 5th takes turns operation;
ComMB～along the macro blocks of path indication 902 samplings;
CurMB～current macro;
ISP～similar macro blocks is searched starting point;
The reference macroblock in Left～current macro CurMB left side;
BestMatchedMB～similar macro blocks;
The median vector estimation design of Median-of-4～tetra-input;
The median vector estimation design of Median-of-6～six input;
MV_L16 * 16, MV_L_R8 * 16, MV_L_D16 * 8, MV_L_U16 * 8, MV_U16 * 16, MV_U_D16 * 8, MV_UR16 * 16, MV_UR_D16 * 8～with reference to motion-vector;
The reference macroblock of Upper～current macro CurMB top;
The top-right reference macroblock of UpperRight～current macro CurMB.
Embodiment
Fig. 1, with a kind of execution mode of block diagram illustration video compression device, wherein applies inter prediction (inter-prediction) and two kinds of technology of infra-frame prediction (intra-prediction), below specifies inter prediction technology.
As shown in the figure, the information that a mobile detection (motion estimation) module 102 is extracted about a present frame and a reference frame from a memory cell 104, to do motion-vector (motion vector) estimation.The output of mobile detection module 102 can be supplied to a motion compensation (motion compensation) module 106, makes it the information of this reference frame make motion compensation, realizes inter prediction.The predictive frame 108 that this motion compensation module 106 produces will be made comparisons with present frame, difference between the two (residues) 110, after conversion, quantizing, can be sent to network abstraction layer (NAL) through scan sorting (reorder) and entropy coding (entropy encoder) again.Institute's exposure technology is made particular design to this mobile detection module 102, wherein similar macro blocks is searched the special design that is set with of starting point (Initial Search Point, ISP).
Content shown in Fig. 1 can system single chip (Soc) mode realize or be realized by computer system/programs executed by microcontroller, or, also can hardware circuit mode realize being made as chip.
Fig. 2 defines for example a similar macro blocks and searches starting point (ISP).In present frame 202, the macro block (being called current macro) of current wish coding is denoted as CurMB.So-called mobile detection is for seeking out the most similar to current macro CurMB one similar macro blocks BestMatchedMB from a reference frame 204.In the execution mode of regional area search (partial search), need to set a similar macro blocks and search starting point ISP, this similar macro blocks is searched the similar macro blocks seek operations in the initial local Search Area 206 of starting point ISP certainly.If similar macro blocks is searched starting point ISP and set mistake, corresponding local Search Area will not include this similar macro blocks BestMatchedMB, and the effect of mobile detection can be greatly affected.According to exposure technology, the setting that similar macro blocks is searched starting point ISP will be more accurate.
About a current macro CurMB, the similar macro blocks of its mobile detection of Fig. 3 diagram is searched starting point ISP and is set needed information.In figure, for current macro CurMB defines a plurality of reference macroblocks, comprising: the reference macroblock Left that is positioned at current macro CurMB left side; Be positioned at the reference macroblock Upper of current macro CurMB top; And be positioned at the top-right reference macroblock UpperRight of current macro CurMB.
Reference macroblock Left can have multiple block of pixels dividing mode, and its motion-vector (motion vectors) all can be used as the reference motion-vector of this current macro CurMB, comprising:
● reference motion-vector MV_L16 * 16 that the block of pixels of 16 * 16 sizes provides;
● reference motion-vector MV_L_R8 * 16 that the block of pixels of right-hand part 8 * 16 sizes provides;
● reference motion-vector MV_L_D16 * 8 that the block of pixels of Lower Half 16 * 8 sizes provides; And
● reference motion-vector MV_L_U16 * 8 that the block of pixels of the first half 16 * 8 sizes provides.
Reference macroblock Upper can have multiple block of pixels dividing mode, and its motion-vector all can be used as the reference motion-vector of this current macro CurMB, comprising:
● reference motion-vector MV_U16 * 16 that the block of pixels of 16 * 16 sizes provides; And
● reference motion-vector MV_U_D16 * 8 that the block of pixels of Lower Half 16 * 8 sizes provides.
● reference macroblock UpperRight can have multiple block of pixels dividing mode, and its motion-vector all can be used as the reference motion-vector of this current macro CurMB, comprising:
● reference motion-vector MV_UR16 * 16 that the block of pixels of 16 * 16 sizes provides; And
● reference motion-vector MV_UR_D16 * 8 that the block of pixels of Lower Half 16 * 8 sizes provides.
With reference to motion-vector, also can be temporarily stored in Fig. 1 memory cell 104 for mobile detection module 102 and extract above.
Different in the position of a present frame according to current macro CurMB, the reference motion-vector quantity that its mobile detection can reference may have change.Referring to Fig. 4, illustrate.
Situation one, current macro CurMB is first macro block 402 of the present frame upper left corner.Now, the motion-vector of current macro CurMB (motion vector) can be set as 0 value.
Situation two, current macro CurMB is positioned at the upper edge area 404 of present frame.Now, current macro CurMB can left side macro block (macro block that Fig. 3 label Left illustrates) macro block for referencial use.A kind of execution mode be make this current macro CurMB mobile detection can with reference to Fig. 3 defined four with reference to motion-vector: MV_L16 * 16, MV_L_R8 * 16, MV_L_D16 * 8 and MV_L_U16 * 8.
Situation three, current macro CurMB is positioned at the left edge area 406 of present frame.Now, current macro CurMB can its top and top-right macro block (macro block that Fig. 3 label Upper and UpperRight illustrate) macro block for referencial use.A kind of execution mode be make this current macro CurMB mobile detection can with reference to Fig. 3 defined four with reference to motion-vector: MV_U16 * 16, MV_U_D16 * 8, MV_UR16 * 16 and MV_UR_D16 * 8.
Situation four, current macro CurMB is positioned at the right edge area 408 of present frame.Now, current macro CurMB can its left side and macro block (macro block that Fig. 3 label Left and the Upper illustrate) macro block for referencial use of top.A kind of execution mode be make this current macro CurMB mobile detection can with reference to Fig. 3 defined four with reference to motion-vector: MV_L16 * 16, MV_L_R8 * 16, MV_U16 * 16 and MV_U_D16 * 8.
Situation five, current macro CurMB is positioned at the non-edge area 410 of present frame.Now, current macro CurMB can its left side, top and top-right macro block (macro block that Fig. 3 label Left, Upper and UpperRight illustrate) macro block for referencial use.A kind of execution mode be make this current macro CurMB mobile detection can with reference to Fig. 3 defined six with reference to motion-vector: MV_L16 * 16, MV_L_R8 * 16, MV_L_D16 * 8, MV_U16 * 16, MV_U_D16 * 8 and MV_UR16 * 16.
Above introduced reference motion-vector design will be more accurate by the setting that makes similar macro blocks search starting point ISP.Special instruction, above narration is not intended to limit the reference motion-vector quantity of each situation or adopts in detail situation.Every in the reference macroblock of this current macro the block of pixels of at least two kinds of sizes obtain with reference to the technology of motion-vector all related to the present invention.
Fig. 5 arranges in disclosed image compressing method the mobile detection about a current macro (as Fig. 2 label CurMB) with flow chart.Step S502, extracts a plurality of with reference to motion-vector of this current macro; These with reference to motion-vector by the block of pixels of at least two kinds of sizes in the reference macroblock of this current macro provided-can have multiple source with reference to aforementioned multiple situation.Step S504, picks out a median motion-vector from these with reference to motion-vector, and according to this median motion-vector, sets a similar macro blocks and search starting point (as Fig. 2 label ISP).Step S506, searches starting point (ISP) in similar macro blocks (as Fig. 2 label BestMatchedMB) of this current macro of reference frame initial search according to this similar macro blocks.Step S508, tries to achieve the motion-vector (motion vector) of this current macro, for video compression according to this similar macro blocks (BestMatchedMB) and this current macro searching.
Special declaration, above similar macro blocks is searched starting point (ISP) and is not limited the regional area search technique being applied in shown in Fig. 2.Hexagonal search (hexagonal search) shown in rhombus search (diamond search) shown in three step searches shown in Fig. 6 A (three step search), Fig. 6 B, Fig. 6 C also can adopt disclosed technology to set its similar macro blocks and search starting point ISP.
About the cited embodiment of above situation two～situation four, the reference motion-vector quantity of a current macro reference is 4.Fig. 7 diagram one median vector estimation design Median-of-4, realizes Fig. 5 step S504-for the cited embodiment of situation two～situation four and with reference to motion-vector, picks out median motion-vector from four.
Median vector estimation design Median-of-4 comprises two comparator cmp1 and cmp2, wherein recycles this two comparator cmp1 and cmp2, to complete the size sequence of a plurality of motion-vectors.Each comparator (cmp1 or cmp2) is responsible for the relatively size of two inputs, and with one first output output the greater and with one second output output smaller.As shown in Figure 7, in first round operation Cycle1, comparator cmp1 picks out the greater Max (A in two input A and B, B) with smaller Min (A, B), and comparator cmp2 picks out the greater Max (C in two input C and D, D) with smaller Min (C, D).Second takes turns operation Cycle2 can reuse comparator cmp1 and cmp2: numerical value Max (A, B) and numerical value Max (C, D) make comparisons again, try to achieve maximum person Max in numerical value A～D (A, B, C, D) and one first median person 702; And numerical value Min (A, B) and numerical value Min (C, D) make comparisons again, try to achieve minimum value person Min in numerical value A～D (A, B, C, D) and one second median person 704.This first and this second median 702 and 704 corresponding two vectors with reference to motion-vector on average can be regarded as disclose among place value motion-vector-with reference to step S504, search starting point (ISP) for setting above-mentioned similar macro blocks.
About the cited embodiment of above situation five, the reference motion-vector quantity of current macro reference is 6.Fig. 8 diagram one median vector estimation design Median-of-6, picks out median motion-vector from six with reference to motion-vector for realizing Fig. 5 step S504-for the cited embodiment of situation five.
Median vector estimation design Median-of-6 recycles the size sequence that two comparator cmp1 and cmp2 complete a plurality of motion-vectors.In the first round and second, take turns in operation Cycle1 and Cycle2, comparator cmp1 and cmp2 adopt the median vector estimation design Median-of-4 operation of Fig. 7, pick out maximum person Max (A:D), minimum value person Min (A:D) and two median persons 802,804 in numerical value A～D.Third round operation Cycle3 reuses comparator cmp1 and cmp2: numerical value Max (A:D) and numerical value E comparison, draw the greater Max (A:D, E) and smaller 806; And numerical value Min (A:D) and numerical value F comparison, draw smaller Min (A:D, F) and the greater 808.Then, fourth round and the 5th is taken turns operation Cycle4 and Cycle5 and is reused median vector estimation design Median-of-4, picks out namely middle two big or small numerical value between six numerical value A～F of two medians 810 in input 802,804,806 and 808 and 812-.Two vectors with reference to motion-vector that median 810 is corresponding with 812 on average can be regarded as disclosed median motion-vector.With reference to step S504, for setting above-mentioned similar macro blocks, search starting point (ISP).
Fig. 7 and Fig. 8 are not intended to limit median vector and estimate in the design shown in figure.Any design that is able to identification median vector all can be applied mechanically.
In addition, the quantity with reference to motion-vector does not limit the above execution mode with source.Take quantity below as example.About the reference motion-vector of negligible amounts-as 3, user can make of median vector estimation design Median-of-4 of Fig. 7 be input as maximum value or minimum value, so that this design Median-of-4 is modified to three inputs are sorted.Or if be 5 with reference to motion-vector number, user can make the median vector of Fig. 8 estimate that one of design Median-of-6 is input as maximum value or minimum value, so that this design Median-of-6 is modified to five inputs are sorted.
A kind of method for searching of Fig. 9 diagram, seeks out similar macro blocks BestMatchedMB for the local Search Area 206 at Fig. 2 according to path indication 902.
A definition one evaluation of estimate cost_value, wherein,
cost_value＝SAD+f(QP)·MV，
SAD be pixel information absolute difference and, f (QP) is a function of quantization parameter QP, the bit that MV is motion-vector represents.
Along path, indicate the various sizes of 902 obtained macro block ComMB to divide (for example, 8 * 4 block of pixels in 4x8 block of pixels ,Huo 8th district in 8 * 8 block of pixels ,Huo 8th district in 4x4 block of pixels ,Huo 4th district in 16 * 8 block of pixels of one 16 * 16 block of pixels or the first half or 8 * 16 block of pixels of 16 * 8 block of pixels of Lower Half or left side or 8 * 16 block of pixels of right-hand part, Huo Shi 6th district) and all need to carry out above-mentioned evaluation of estimate cost_value estimation.Be that the division of pin different size provides an exclusive comparator, the motion-vector of the affiliated macro block of the minimum evaluation of estimate of each size cost_value person is noted down in buffer.The minimum evaluation of estimate cost_value of each size can make comparisons each other again; Motion-vector based on storing in the corresponding buffer of minimum person, can seek obtaining similar macro blocks BestMatchedMB.
Although the present invention discloses as above with a plurality of embodiment; so it is not intended to limit the present invention, any those skilled in the art, without departing from the spirit and scope of the present invention; when doing trickle change and modification, so protection scope of the present invention is when being as the criterion depending on the claim person of defining.
Claims (14)
1. a video-frequency compression method, comprising:
Extract a plurality of with reference to motion-vector of a current macro, with reference to motion-vector, the block of pixels by least two kinds of sizes in the reference macroblock of this current macro is provided for these;
From these, with reference to motion-vector, pick out a median motion-vector, and according to this median motion-vector, set a similar macro blocks and search starting point;
According to this similar macro blocks, search starting point in similar macro blocks of this current macro of reference frame initial search; And
According to this similar macro blocks and this current macro of searching, try to achieve the motion-vector of this current macro for video compression.
2. video-frequency compression method according to claim 1, wherein, when this current macro is positioned at the upper edge area of a present frame, the reference macroblock of this current macro is positioned at the left of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, and the block of pixels of two 16 * 8 sizes provides one the 3rd and one the 4th with reference to motion-vector, the block of pixels of this 8 * 16 size is close to this current macro, and the block of pixels of above-mentioned two 16 * 8 sizes forms this reference macroblock.
3. video-frequency compression method according to claim 1, wherein, when this current macro is positioned at the left edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock and one second reference macroblock, this the first reference macroblock is positioned at the top of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one second with reference to motion-vector, this the second reference macroblock is positioned at the upper right side of this current macro, and be to provide one the 3rd with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 4th with reference to motion-vector, the block of pixels of this 16 * 8 size in this first reference macroblock is close to this current macro, and the block of pixels of this 16 * 8 size in this second reference macroblock belongs to the Lower Half of this second reference macroblock.
4. video-frequency compression method according to claim 1, wherein, when this current macro is positioned at the right edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock and one second reference macroblock, this the first reference macroblock is positioned at the left side of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, this the second reference macroblock is positioned at the top of this current macro, and be to provide one the 3rd with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 4th with reference to motion-vector, the block of pixels of this 8 * 16 size in this first reference macroblock is close to this current macro, and the block of pixels of this 16 * 8 size in this second reference macroblock is close to this current macro.
5. video-frequency compression method according to claim 1, wherein, when this current macro is positioned at the non-edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock, one second reference macroblock and one the 3rd reference macroblock, this the first reference macroblock is positioned at the left side of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, and the block of pixels of one 16 * 8 sizes provides one the 3rd with reference to motion-vector, this the second reference macroblock is positioned at the top of this current macro, and be to provide one the 4th with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 5th with reference to motion-vector, the 3rd reference macroblock is positioned at the upper right side of this current macro, and be to provide one the 6th with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of this 8 * 16 size in this first reference macroblock is close to this current macro, the block of pixels of this 16 * 8 size in this first reference macroblock belongs to the Lower Half of this first reference macroblock, and the block of pixels of this 16 * 8 size in this second reference macroblock is close to this current macro.
6. a video compression device, comprising:
One memory cell, a current macro a plurality of with reference to motion-vector that store a reference frame and a present frame, with reference to motion-vector, the block of pixels by least two kinds of sizes in the reference macroblock of this current macro is provided for these; And
One mobile detection module, these that extract this current macro from this memory cell are with reference to motion-vector, and from these, with reference to motion-vector, pick out a median motion-vector and search starting point to set a similar macro blocks according to this median motion-vector, and according to this similar macro blocks, search similar macro blocks of this this current macro of reference frame initial search that starting point stores in this memory cell, and the motion-vector of trying to achieve this current macro according to this similar macro blocks searching and this current macro is for video compression.
7. video compression device according to claim 6, wherein, when this current macro is positioned at the upper edge area of a present frame, the reference macroblock of this current macro is positioned at the left of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, and the block of pixels of two 16 * 8 sizes provides one the 3rd and one the 4th with reference to motion-vector, the block of pixels of this 8 * 16 size is close to this current macro, and the block of pixels of above-mentioned two 16 * 8 sizes forms this reference macroblock.
8. video compression device according to claim 7, also comprises:
One first comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
One second comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
Wherein:
In a first round operation, above-mentioned first to fourth with reference to motion-vector be supplied between two this first and this second comparator make comparisons;
In one second, take turns operation, this first and this second comparator be again used, therebetween, the signal that this first output of this first comparator is exported first round when this operation be the signal exported when this operation with this first output of this second comparator first round make comparisons to pick out this first to the 4th with reference to motion-vector in maximum person and one first median person, and the signal that this second output of this first comparator is exported first round when this operation be the signal exported when this operation with this second output of this second comparator first round make comparisons to pick out this first to the 4th with reference to motion-vector in minimum value person and one second median person, and
This first to the 4th with reference in motion-vector through be recognized as this first with this second median person's vector be on average above-mentioned median motion-vector.
9. video compression device according to claim 6, wherein, when this current macro is positioned at the left edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock and one second reference macroblock, this the first reference macroblock is positioned at the top of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one second with reference to motion-vector, this the second reference macroblock is positioned at the upper right side of this current macro, and be to provide one the 3rd with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 4th with reference to motion-vector, the block of pixels of this 16 * 8 size in this first reference macroblock is close to this current macro, and the block of pixels of this 16 * 8 size in this second reference macroblock belongs to the Lower Half of this second reference macroblock.
10. video compression device according to claim 9, also comprises:
One first comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
One second comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
Wherein:
In a first round operation, above-mentioned first to fourth with reference to motion-vector be supplied between two this first and this second comparator make comparisons;
In one second, take turns operation, this first and this second comparator be again used, therebetween, the signal that the signal that this first output of this first comparator is exported first round when this operation and this first output of this second comparator are exported first round when this operation make comparisons to pick out this first to the 4th with reference to motion-vector in maximum person and one first median person, and the signal that the signal that this second output of this first comparator is exported first round when this operation and this second output of this second comparator are exported first round when this operation make comparisons to pick out this first to the 4th with reference to motion-vector in minimum value person and one second median person, and
This first to the 4th with reference in motion-vector through be recognized as this first with this second median person's vector be on average above-mentioned median motion-vector.
11. video compression devices according to claim 6, wherein, when this current macro is positioned at the right edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock and one second reference macroblock, this the first reference macroblock is positioned at the left side of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, this the second reference macroblock is positioned at the top of this current macro, and be to provide one the 3rd with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 4th with reference to motion-vector, the block of pixels of this 8 * 16 size in this first reference macroblock is close to this current macro, and the block of pixels of this 16 * 8 size in this second reference macroblock is close to this current macro.
12. video compression devices according to claim 11, also comprise:
One first comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
One second comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
Wherein:
In a first round operation, above-mentioned first to fourth with reference to motion-vector be supplied between two this first and this second comparator make comparisons;
In one second, take turns operation, this first and this second comparator be again used, therebetween, the signal that this first output of this first comparator is exported first round when this operation be the signal exported when this operation with this first output of this second comparator first round make comparisons to pick out this first to the 4th with reference to motion-vector in maximum person and one first median person, and the signal that this second output of this first comparator is exported first round when this operation be the signal exported when this operation with this second output of this second comparator first round make comparisons to pick out this first to the 4th with reference to motion-vector in minimum value person and one second median person, and
This first to the 4th with reference in motion-vector through be recognized as this first with this second median person's vector be on average above-mentioned median motion-vector.
13. video compression devices according to claim 6, wherein, when this current macro is positioned at the non-edge area of a present frame, the reference macroblock of this current macro comprises one first reference macroblock, one second reference macroblock and one the 3rd reference macroblock, this the first reference macroblock is positioned at the left side of this current macro, and be to provide one first with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of one 8 * 16 sizes provides one second with reference to motion-vector, and the block of pixels of one 16 * 8 sizes provides one the 3rd with reference to motion-vector, this the second reference macroblock is positioned at the top of this current macro, and be to provide one the 4th with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, and the block of pixels of one 16 * 8 sizes provides one the 5th with reference to motion-vector, the 3rd reference macroblock is positioned at the upper right side of this current macro, and be to provide one the 6th with reference to motion-vector with the block of pixels of one 16 * 16 sizes wherein, the block of pixels of this 8 * 16 size in this first reference macroblock is close to this current macro, the block of pixels of this 16 * 8 size in this first reference macroblock belongs to the Lower Half of this first reference macroblock, and the block of pixels of this 16 * 8 size in this second reference macroblock is close to this current macro.
14. video compression devices according to claim 13, also comprise:
One first comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
One second comparator, compares the sizes of two inputs, and with one first output output the greater and with one second output output smaller;
Wherein:
In a first round operation, optional above-mentioned first to the 6th with reference to one first, 1 second, 1 the 3rd and one four-way amount in motion-vector be supplied between two this first and this second comparator make comparisons;
In one second, take turns operation, this first and this second comparator be again used, therebetween, the signal that this first output of this first comparator is exported first round when this operation is that the signal exported when this operation with this first output of this second comparator makes comparisons to pick out this first to maximum person and one first median reference in this four-way amount first round, and the signal that this second output of this first comparator is exported first round when this operation is that the signal exported when this operation with this second output of this second comparator makes comparisons to pick out this first to minimum value person and one second median reference in this four-way amount first round,
In a third round operation, this first and this second comparator be again used, therebetween, above-mentioned first to the 6th with reference to one the 5th vector in motion-vector, first to maximum person in this four-way amount, to make comparisons to pick out smaller be one the 3rd median reference with this, and above-mentioned first to the 6th with reference to one the 6th vector in motion-vector, first to minimum value person in this four-way amount, to make comparisons to pick out the greater be one the 4th median reference with this;
In a fourth round and the 5th, take turns operation, this first and twice operation repeatedly of this second comparator, with pick out above-mentioned first to fourth median with reference in one first and one second median person; And
This first to the 4th median with reference in through be recognized as this first with this second median person's vector be on average above-mentioned median motion-vector.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW101124167A TWI499283B (en) | 2012-07-05 | 2012-07-05 | Video compression method and device |
| TW101124167 | 2012-07-05 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103533375A true CN103533375A (en) | 2014-01-22 |
| CN103533375B CN103533375B (en) | 2016-12-21 |
Family
ID=49934985
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210276252.8A Active CN103533375B (en) | 2012-07-05 | 2012-08-06 | Video compression method and video compression device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN103533375B (en) |
| TW (1) | TWI499283B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000014682A2 (en) * | 1998-09-07 | 2000-03-16 | Thomson Multimedia | Method of motion estimation for transmission cost reduction of motion vectors |
| US20030123547A1 (en) * | 2002-01-02 | 2003-07-03 | Samsung Electronics Co., Ltd. | Apparatus of motion estimation and mode decision and method thereof |
| CN1791224A (en) * | 2005-12-19 | 2006-06-21 | 宁波大学 | Self-adaptive block searching range rapid motion estimating method based on H.264 |
| CN1829325A (en) * | 2005-03-01 | 2006-09-06 | 凌阳科技股份有限公司 | Method and system for dynamically adjusting motion estimation |
| US20090245374A1 (en) * | 2008-03-26 | 2009-10-01 | Mediatek Inc. | Video encoder and motion estimation method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2005029833A2 (en) * | 2003-09-21 | 2005-03-31 | Servision Ltd. | Deriving motion detection information from motion-vector-search type video encoders |
| KR100631777B1 (en) * | 2004-03-31 | 2006-10-12 | 삼성전자주식회사 | Method and apparatus for effectively compressing motion vectors in multi-layer |
| TW200727704A (en) * | 2006-01-12 | 2007-07-16 | Advance Multimedia Internet Technology Inc | Inspection method for speeding up dynamic digital image compression by reducing number of prediction modes |
| TWI324484B (en) * | 2006-05-23 | 2010-05-01 | Jiunin Guo | Method for reducing calculation complexity of video compression standard |
| US8488676B2 (en) * | 2007-05-14 | 2013-07-16 | Himax Technologies Limited | Motion estimation method |
| EP3852371B1 (en) * | 2009-07-03 | 2024-03-20 | Orange | Prediction of a movement vector of a current image partition pointing to a reference area that covers several reference image partitions, encoding and decoding using such a prediction |
-
2012
- 2012-07-05 TW TW101124167A patent/TWI499283B/en active
- 2012-08-06 CN CN201210276252.8A patent/CN103533375B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000014682A2 (en) * | 1998-09-07 | 2000-03-16 | Thomson Multimedia | Method of motion estimation for transmission cost reduction of motion vectors |
| US20030123547A1 (en) * | 2002-01-02 | 2003-07-03 | Samsung Electronics Co., Ltd. | Apparatus of motion estimation and mode decision and method thereof |
| CN1829325A (en) * | 2005-03-01 | 2006-09-06 | 凌阳科技股份有限公司 | Method and system for dynamically adjusting motion estimation |
| CN1791224A (en) * | 2005-12-19 | 2006-06-21 | 宁波大学 | Self-adaptive block searching range rapid motion estimating method based on H.264 |
| US20090245374A1 (en) * | 2008-03-26 | 2009-10-01 | Mediatek Inc. | Video encoder and motion estimation method |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201404157A (en) | 2014-01-16 |
| TWI499283B (en) | 2015-09-01 |
| CN103533375B (en) | 2016-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102081213B1 (en) | Image prediction method and related device | |
| AU2020207857B2 (en) | Image encoding method and apparatus, and image decoding method and apparatus | |
| CN110024392A (en) | Low complex degree sign prediction for video coding | |
| TW202002657A (en) | Component-dependent sub-block dividing | |
| CN107690810A (en) | It is determined that the system and method for the illumination compensation state for video coding | |
| CN102801972A (en) | Characteristic-based motion vector estimation and transmission method | |
| JP2015515806A5 (en) | ||
| KR101354014B1 (en) | Coding structure | |
| TW202021344A (en) | Shape dependent intra coding | |
| US10034016B2 (en) | Coding apparatus, computer system, coding method, and computer product | |
| CN106331730B (en) | H.264 video same quantization factor double compression detection method | |
| EP2280550A1 (en) | Mask generation for motion compensation | |
| CN110933415B (en) | Object region detection method, device and non-transitory computer readable medium for object region detection | |
| CN106063268B (en) | Code-transferring method, transcoding device and information storage medium | |
| Michael et al. | Time series classification using compressed recurrence plots | |
| CN103533375A (en) | Video compression method and video compression device | |
| CN110800301A (en) | Control method and device of coding equipment and storage medium | |
| KR102345258B1 (en) | Object Region Detection Method, Device and Computer Program Thereof | |
| Loose et al. | A hardware-friendly and configurable heuristic targeting vvc inter-frame prediction | |
| Zhang et al. | Feature-matching based motion prediction for high efficiency video coding in cloud | |
| CN106162196B (en) | Intelligent analysis-oriented video coding system and method | |
| JP6468703B2 (en) | Motion detection device, motion detection method, and program | |
| CN113556551B (en) | Encoding and decoding method, device and equipment | |
| RU2776941C1 (en) | Video encoding and decoding | |
| RU2793275C2 (en) | Video coding and decoding |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |