CN102111620B - Frame rate downsampling transcoding method and device - Google Patents

Abstract

The invention discloses a frame rate downsampling transcoding method and device. When existing video frames in a part of video streams which are decoded are forward predictive frames and the forward predictive frames are reserved, for any interframe coding macroblock in the existing video frames, under the condition that one or more than one interframe coding macroblocks in the interframe coding macroblocks have original motion vector information and/or reconfigured motion vector information, the original motion vector information and/or the reconfigured motion vector information of one or more than one interframe coding macroblocks can be used to acquire rebuilt motion vector information of the interframe coding macroblocks which do not have the original motion vector information or the reconfigured motion vector information in interframe coding macroblocks.

Description

Frame rate downsampling transcoding method and device

Technical field

The present invention relates to image processing field, relate more specifically to a kind of frame-skipping code-transferring method and device.

Background technology

Along with the high speed development of the Internet and mobile communication technology, the various client computer with different performance have appearred, as mobile phone, palmtop PC, laptop computer etc., these client computer an urgent demands can the wireless access the Internet, shares the resource on the Internet.Because the Internet and wireless network have different bandwidth, thereby also corresponding different transmission code rate just.If directly by wireless network, send the video flowing compressed to client computer from the Internet, will there will be the situation of video flowing and transmission channel mismatch.Now, just need between the Internet and client computer, arrange and act on behalf of transducer, the video flowing compressed is carried out to the code check conversion, the high-speed video circulation of having compressed is changed into to the video flowing of low code check, to guarantee the correct transmission of video flowing in mobile wireless network, for the mobile subscriber provides the Video service of different service quality.

Can reduce by the frame per second that reduces video flowing the code check of video flowing, and in order to reduce the frame per second of video flowing, consideration directly abandons non-reference frame in video flowing to avoid the impact of frame subsequently usually.This is the simplest situation, and the processing of this frame-skipping transcoding can not change the referring-to relation between frame of video.Yet, in the time must abandoning reference frame, motion vector (Motion Vector, MV) is rebuild will become a subject matter.

Video encoding standard H.264 in, can there is different sizes as the encoding block of coding unit.Wherein, the size of Intra-coded blocks can be 16x16,8x8 or 4x4, and the size of inter-coded block can be 16x16,16x8,8x16 or 8x8.And inter-coded block can be divided into a plurality of sub-blocks, for example, the inter-coded block of 8x8 can be further divided into the sub-block of two 8x4 or 4x8 or the sub-block of four 4x4.Although must adopt identical coding mode (comprising intraframe coding and interframe encoding mode) to same macro block (Macro Block, MB) the different coding piece in is encoded, but the different inter-coded block in same inter-coded macroblocks (for example, encoding block A and encoding block B) can have the reference block in different reference frames respectively.That is to say, a frame of video can have a plurality of reference frames, and an inter-coded macroblocks can have a plurality of reference blocks that are arranged in different reference frames, and the reference block of the different sub-blocks in same 8x8 inter-coded block must be in same reference frame.

When the video flowing that has a plurality of reference frames when each frame of video to wherein carries out the frame-skipping transcoding, because the one or more reference frames that comprise reference block are skipped, make one or more original motion vectors of the one or more inter-coded macroblocks in present frame no longer valid.Fig. 1 show the down-sampling rate be 3: 2 (, every next but two frame of video is skipped a frame of video) situation under, the schematic diagram of the motion vector of each inter-coded block in the inter-coded macroblocks that comprises for example A, B, C, tetra-inter-coded block of D in present frame N.As shown in fig. 1, in eight frame of video of N-7 to N, frame of video N-5 and N-2 are skipped, and the reference block A ' of the inter-coded block A in present frame N is arranged in frame of video N-4, reference block B ' and the D ' of inter-coded block B and D are arranged in the frame of video N-2 be skipped, and the reference block C ' of inter-coded block C is arranged in frame of video N-2.Wherein, because the frame of video at the reference block place of inter-coded block B and D is skipped, so the original motion vector of inter-coded block B and D is no longer valid.In this case, if again the inter-coded macroblocks at inter-coded block A, B, C and D place is carried out to intraframe coding, can greatly increase bit rate; But, because the original motion vector of two inter-coded block in this inter-coded macroblocks is invalid, cause again carrying out interframe encode to this inter-coded macroblocks.

Summary of the invention

One or more problems in view of the above, the invention provides a kind of frame rate downsampling transcoding method and device.

Frame rate downsampling transcoding method according to the embodiment of the present invention, comprise: received compressing video frequency flow is carried out to partial decoding of h, and obtain the original coding information of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, partial decoding of h and the video flowing after partial decoding of h; current video frame in video flowing after partial decoding of h is forward predicted frame and will be retained the time, for any one inter-coded macroblocks in current video frame, in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block and/or the motion vector information reconfigured, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks, wherein, in described inter-coded macroblocks, if the reference frame of an inter-coded block will be retained, this inter-coded block has the original motion Vector Message, if the reference frame of an inter-coded block will be skipped, but this inter-coded block can reconfigure out at a described original motion Vector Message by one or more encoding blocks that on the frame of video be retained, the institute overlay area relates to by the reference block on its reference frame by this inter-coded block with respect to an approximate motion vector information by the frame of video that is retained, this inter-coded block has the motion vector information reconfigured, and utilize the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the motion vector information of the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after partial decoding of h/reconfigure/reconstructed motion vector information and residual information, partial decoding of h and the original coding information of each intra-coded frame in the video flowing after partial decoding of h, the video flowing after partial decoding of h is carried out to the frame rate downsampling transcoding.

Frame rate downsampling transcoding device according to the embodiment of the present invention, comprise: the partial decoding of h unit, be configured to received compressing video frequency flow is carried out to partial decoding of h, and obtain the original coding information of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, partial decoding of h and the video flowing after partial decoding of h, the vector reconstruction unit, be configured to current video frame in the video flowing after partial decoding of h and be forward predicted frame and will be retained the time, for any one inter-coded macroblocks in current video frame, in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block and/or the motion vector information reconfigured, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks, wherein, in described inter-coded macroblocks, if the reference frame of an inter-coded block will be retained, this inter-coded block has the original motion Vector Message, if the reference frame of an inter-coded block will be skipped, but this inter-coded block can reconfigure out at a described original motion Vector Message by one or more encoding blocks that on the frame of video be retained, the institute overlay area relates to by the reference block on its reference frame by this inter-coded block with respect to an approximate motion vector information by the frame of video that is retained, this inter-coded block has the motion vector information reconfigured, and down-sampling transcoding units, be configured to utilize the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the motion vector information of the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after partial decoding of h/reconfigure/reconstructed motion vector information and residual information, partial decoding of h and the original coding information of each intra-coded frame in the video flowing after partial decoding of h, the video flowing after partial decoding of h is carried out to the frame rate downsampling transcoding.

In the present invention, can there is the original motion Vector Message of inter-coded block of original motion Vector Message or the motion vector information reconfigured with inter-coded block of the motion vector information reconfigured by utilization, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured, thereby the inter-coded block that can in the down-sampling transcoding process, to neither having the original motion Vector Message, not have again a motion vector information reconfigured is carried out interframe encode.

The accompanying drawing explanation

From the description to the specific embodiment of the present invention below in conjunction with accompanying drawing, the present invention may be better understood, wherein:

Fig. 1 shows in the situation that the down-sampling rate is 3: 2, the schematic diagram of the motion vector of each inter-coded block in the inter-coded macroblocks that comprises for example A, B, C, tetra-inter-coded block of D in present frame N;

Fig. 2 shows the flow chart according to the frame rate downsampling transcoding method of the embodiment of the present invention;

Fig. 3 shows the block diagram according to the frame rate downsampling transcoding device of the embodiment of the present invention;

Fig. 4 shows the schematic diagram of processing that the original motion Vector Message that utilizes an inter-coded block according to the embodiment of the present invention obtains the reconstructed motion vector information of another inter-coded block;

Fig. 5 shows the schematic diagram of processing that the motion vector information reconfigured that utilizes an inter-coded block according to the embodiment of the present invention obtains the reconstructed motion vector information of another inter-coded block;

Fig. 6 shows in the situation that exist an inter-coded block with original motion Vector Message and one to have the motion vector information reconfigured in comprising the inter-coded macroblocks of four inter-coded block, obtains the schematic diagram of the processing of the reconstructed motion vector information that neither has the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured; And

Fig. 7 shows in the situation that the schematic diagram of the processing that the down-sampling rate is the motion vector reconfigured that for example obtains inter-coded block A at 2: 1.

Embodiment

Below will describe feature and the exemplary embodiment of various aspects of the present invention in detail.Many details have been contained in following description, in order to complete understanding of the present invention is provided.But, it will be apparent to one skilled in the art that the present invention can in the situation that some details in not needing these details implement.To the description of embodiment, be below only in order by example of the present invention is shown, to provide the clearer understanding to the present invention.The present invention never is limited to following any concrete configuration and the algorithm proposed, but has covered under the premise of without departing from the spirit of the present invention any modification, replacement and the improvement of coherent element, parts and algorithm.

Fig. 2 shows the flow chart according to the frame rate downsampling transcoding method of the embodiment of the present invention.As shown in Figure 2, this frame rate downsampling transcoding method comprises:

S202, received compressing video frequency flow is carried out to partial decoding of h, and obtain the original coding information of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, partial decoding of h and the video flowing after partial decoding of h;

S204, current video frame in video flowing after partial decoding of h is forward predicted frame and will be retained the time, for any one inter-coded macroblocks in current video frame, in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block and/or the motion vector information reconfigured, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks, and

S206, utilize the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the motion vector information of the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after partial decoding of h/reconfigure/reconstructed motion vector information and residual information, partial decoding of h and the original coding information of each intra-coded frame in the video flowing after partial decoding of h, the video flowing after partial decoding of h is carried out to the frame rate downsampling transcoding.

Fig. 3 shows the block diagram according to the frame rate downsampling transcoding device of the embodiment of the present invention.As shown in Figure 3, this frame rate downsampling transcoding device comprises partial decoding of h unit 302, vector reconstruction unit 304 and down-sampling transcoding units 306.

Wherein, 302 pairs of the partial decoding of h unit compressing video frequency flow received carries out partial decoding of h, and obtain the original coding information (that is, execution step S202) of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, partial decoding of h and the video flowing after partial decoding of h.Current video frame in video flowing after partial decoding of h is forward predicted frame and will be retained the time, for any one inter-coded macroblocks in current video frame, vector reconstruction unit 304 is in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block or the motion vector information reconfigured, obtain the reconstructed motion vector information that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured in described inter-coded macroblocks (, execution step S204).Down-sampling transcoding units 306 utilizes the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the motion vector information of the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after partial decoding of h/reconfigure/reconstructed motion vector information and residual information, partial decoding of h and the original coding information of each intra-coded frame in the video flowing after partial decoding of h, video flowing after partial decoding of h is carried out to frame rate downsampling transcoding (that is, execution step S206).

In video flowing after partial decoding of h, if the reference frame of an inter-coded block will be retained, this inter-coded block has the original motion Vector Message; If the reference frame of an inter-coded block will be skipped, but this inter-coded block can reconfigure out at a described original motion Vector Message by one or more encoding blocks that on the frame of video be retained, the institute overlay area relates to by the reference block on its reference frame by this inter-coded block with respect to an approximate motion vector information by the frame of video that is retained, and this inter-coded block has the motion vector information reconfigured.In addition, for each frame of video in the video flowing after partial decoding of h, for example can judge that this frame of video will be retained or will be skipped in the frame rate downsampling transcoding process according to predetermined down-sampling frame per second.

Fig. 4 shows the schematic diagram of processing that the original motion Vector Message that utilizes an inter-coded block according to the embodiment of the present invention obtains the reconstructed motion vector information of another inter-coded block.As shown in Figure 4, an inter-coded macroblocks in present frame N is comprised of inter-coded block A and the B of 2 8x16.Wherein, because frame of video (that is, the reference frame of the inter-coded block A) N-m at the reference block A ' place of inter-coded block A will be retained, so its original motion Vector Message can be used in during frame rate downsampling transcoding subsequently processes; Because frame of video (that is, the reference frame of the inter-coded block B) N-n at the reference block B ' place of inter-coded block B will be skipped, so its original motion Vector Message can not be used in during frame rate downsampling transcoding subsequently processes.

Can carry out interframe encode for the frame rate downsampling transcoding subsequently to inter-coded block B in processing, need to find out inter-coded block B with respect to a motion vector information by the frame of video that is retained.In the situation that, shown in Fig. 4, vector reconstruction unit 304 for example can obtain by following processing the reconstructed motion vector of inter-coded block B:

At first, select the new reference frame using the frame of video N-n-r that is retained as inter-coded block B before the reference frame N-n of inter-coded block B, that approach this reference frame N-n most;

Then, according to following equation (1) calculate inter-coded block B with respect to it reference block B in frame of video N-n-r ^*motion vector (that is, the reconstructed motion vector information of inter-coded block B):

${\mathrm{MV}}_{B^{*}-B}={\mathrm{MV}}_{A^{\′}-A}*\frac{N-(N-n-r)}{N-(N-m)}---\left(1\right)$

Wherein, MV _{a '-A}mean the original motion vector of inter-coded block A with respect to frame of video N-m,

mean the reconstructed motion vector of inter-coded block B with respect to frame of video N-n-r.

Fig. 5 shows the schematic diagram of processing that the motion vector information reconfigured that utilizes an inter-coded block according to the embodiment of the present invention obtains the reconstructed motion vector information of another inter-coded block.As shown in Figure 5, an inter-coded macroblocks in present frame N is comprised of inter-coded block A and the B of 2 8x16.Because the reference block A ' of inter-coded block A and B and the frame of video at B ' place (that is, the reference frame of inter-coded block A and B) N-k and N-n will be skipped, so their original motion Vector Message can not be used in during frame rate downsampling transcoding subsequently processes.

But, the regional A covered on the frame of video N-m by being retained according to the reference block A ' of inter-coded block A ^*the weighted sum of the original motion vector of the encoding block related to, can show that inter-coded block A is with respect to the regional A on frame of video N-m ^*motion vector (that is, the motion vector information reconfigured of inter-coded block A), thereby can in frame rate downsampling transcoding is subsequently processed, utilize the motion vector reconfigured of inter-coded block A to carry out interframe encode to inter-coded block A.And can not draw by the above processing that is similar to inter-coded block A the motion vector information reconfigured of inter-coded block B.

Can carry out interframe encode for the frame rate downsampling transcoding subsequently to inter-coded block B in processing, need to find out inter-coded block B with respect to a motion vector information by the frame of video that is retained.In the situation that, shown in Fig. 5, vector reconstruction unit 304 for example can obtain by following processing the reconstructed motion vector of inter-coded block B:

At first, select the new reference frame using the frame of video N-n-r that is retained as inter-coded block B before the reference frame N-n of inter-coded block B, that approach this reference frame N-n most;

Then, according to following equation (2) calculate inter-coded block B with respect to it reference block B in frame of video N-n-r ^*motion vector (that is, the reconstructed motion vector information of inter-coded block B):

${\mathrm{MV}}_{B^{*}-B}={\mathrm{MV}}_{A^{*}-A}*\frac{N-(N-n-r)}{N-(N-m)}---\left(2\right)$

Wherein,

mean the motion vector that reconfigure of inter-coded block A with respect to frame of video N-m,

mean the reconstructed motion vector of inter-coded block B with respect to frame of video N-n-r.

Fig. 6 shows in the situation that exist an inter-coded block with original motion Vector Message and one to have the motion vector information reconfigured in comprising the inter-coded macroblocks of four inter-coded block, obtains the schematic diagram of the processing of the reconstructed motion vector information that neither has the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured.

As shown in Figure 6, an inter-coded macroblocks in present frame N is comprised of inter-coded block A, B, C and the D of 4 4x4.Reference block A due to inter-coded block A and B ^*and B ^*the frame of video at place (that is, the reference frame of inter-coded block A and B) N-k and N-n will be retained, so their original motion Vector Message can be used in frame rate downsampling transcoding processing subsequently; Because the reference block C ' of inter-coded block C and D and the frame of video at D ' place (that is, the reference frame of inter-coded block C and D) N-m and N-r will be skipped, so their original motion Vector Message can not be used in during frame rate downsampling transcoding subsequently processes.

Can carry out interframe encode for the frame rate downsampling transcoding subsequently to inter-coded block C and D in processing, need to find out inter-coded block C and D with respect to a motion vector information by the frame of video that is retained.In the situation that, shown in Fig. 6, vector reconstruction unit 304 can obtain for example reconstructed motion vector of inter-coded block C by following processing:

At first, the original motion vector of the inter-coded block B that its reference frame in selection inter-coded block A and B is nearer apart from the reference frame N-m of inter-coded block C is as the reference motion vector of the reconstructed motion vector that solves inter-coded block C, here the original motion vector of inter-coded block B is designated as to refMV, the reference frame of inter-coded block B is designated as to refList _ref;

Then, select the new reference frame using the frame of video N-m-a that is retained as inter-coded block C before the reference frame N-m of inter-coded block C, that approach this reference frame N-m most, this new reference frame is designated as to refList here;

Then, according to following equation (3) calculate inter-coded block C with respect to it reference block C in frame of video N-m-a ^*motion vector

(that is, the reconstructed motion vector information of inter-coded block C):

${\mathrm{MV}}_{C-C^{*}}=\mathrm{refMV}*\frac{N-\mathrm{refList}}{N-\mathrm{refLis}{t}_{\mathrm{ref}}}---\left(3\right)$

In the situation that the inter-coded block A in Fig. 6 and B have the motion vector information reconfigured, also can utilize said method to solve the reconstructed motion vector of inter-coded block C, what only refMV meaned is the motion vector reconfigured of inter-coded block B.Can utilize said method to solve the reconstructed motion vector of inter-coded block D.

The above method of describing in conjunction with Fig. 6 also can promote the use of that inter-coded block A, B and C have the original motion Vector Message simultaneously or the situation of the motion vector information that reconfigures under.In this case, at first need to select the original motion vector of its reference frame in inter-coded block A, B and C inter-coded block C nearer apart from the reference frame N-r of inter-coded block D or the motion vector that reconfigures as the reference motion vector of the reconstructed motion vector that solves inter-coded block D.

The motion vector information that a plurality of inter-coded block in Fig. 6 have the original motion Vector Message and/or reconfigure, for example inter-coded block A has the original motion Vector Message and inter-coded block B and C have in the situation of the motion vector information reconfigured, and can utilize the original motion Vector Message of inter-coded block A to obtain the reconstructed motion vector information of inter-coded block D by the method for describing in conjunction with Fig. 4.

The motion vector information that a plurality of inter-coded block in Fig. 6 have the original motion Vector Message and/or reconfigure, for example inter-coded block A and B have the original motion Vector Message and inter-coded block C has in the situation of the motion vector information reconfigured, and can utilize the original motion Vector Message of inter-coded block A and B to obtain the reconstructed motion vector information of inter-coded block D by the method for describing in conjunction with Fig. 6.

The processing of the motion vector reconfigured that obtains an inter-coded block is described below in conjunction with Fig. 7.Fig. 7 shows in the situation that the schematic diagram of the processing that the down-sampling rate is the motion vector reconfigured that for example obtains inter-coded block A at 2: 1.As shown in Figure 7, the inter-coded block A in present frame N is of a size of:

S＝h×v (4)

In equation (4), the value of h and v can be 4,8 or 16.The reference block A ' (it comprise from inter-coded block A equal number of pixels) of inter-coded block A in the frame of video N-1 by being skipped covered and has been positioned at different macro blocks (Macro Block in frame of video N-1, MB) a plurality of encoding blocks in (here using 8 as example, describe, certainly, these encoding blocks also may be arranged in same macro block).Here the capped a plurality of encoding blocks of hypothesis are inter-coded block, and these encoding blocks have respectively the motion vector that related blocks from frame of video N-2 is pointed to it.Because frame of video N-1 will be skipped, so frame of video N-1 is no longer valid to the referring-to relation of frame of video N-1 to referring-to relation and the frame of video N-2 of frame of video N.Yet, for the video flowing to after partial decoding of h carries out recompile, need the referring-to relation to frame of video N of each encoding block in reconstruction video frame N-2.In Fig. 7, reference block A ' has covered 8 encoding blocks in frame of video N-1.Each encoding block i (i=1,2,3 ..., 8) all there is the motion vector MVi with respect to the related blocks in frame of video N-2, the horizontal component of this motion vector MVi is MV (x _i), vertical component is MV (y _i).The area in the zone that referenced A ' of encoding block i covers is A (i).The area in the zone that 1～8 referenced A ' of the encoding block in Fig. 7 covers is respectively:

A(1)＝a×b (5)

A(2)＝b×c (6)

A(3)＝g×(a+c) (7)

A(4)＝g×(d+e) (8)

A(5)＝b×d (9)

A(6)＝b×e (10)

A(7)＝b×f (11)

A(8)＝g×f (12)

Wherein,

b+g＝h (13)

a+c+d+e+f＝v (14)

Can be according to following equation, approximate calculation reference block A ' is with respect to the motion vector MV ' of the related blocks in frame of video N-2:

${\mathrm{MV}}^{\′}\≈\frac{1}{S}\underset{i=1}{\overset{8}{\mathrm{\Σ}}}\mathrm{MVi}\×A\left(i\right)---\left(15\right)$

Then, can be according to following equation, utilize inter-coded block A with respect to the motion vector MV of reference block A ' and reference block A ' with respect to the motion vector MV ' calculating inter-coded block A of the related blocks in frame of video N-2 the motion vector MVnew with respect to the related blocks in frame of video N-2:

MVnew＝MV+MV′ (16)

Accurate as far as possible for the motion vector MVnew that makes inter-coded block A, calculating prediction piece A ' with respect to the motion vector MV ' of the related blocks in frame of video N-2 before, at first the variance of the motion vector of calculation code piece 1～8.If the variance of the motion vector of encoding block 1～8 is very large, mean that the activity difference of these encoding blocks is also very large, so the correlation between these pieces is very little.In this case, in transform coding process subsequently, should carry out intraframe coding rather than interframe encode to inter-coded block A, so no longer need to calculate the motion vector MVnew of inter-coded block A.For example, the condition of inter-coded block A being carried out to intraframe coding can be expressed as follows:

InterToIntraFlag＝

(var(MV(x ₀)，MV(x ₁)，…，MV(x _i))＞T)&&(var(MV(y ₀)，MV(y ₁)，…，MV(y _i))＞T)

(14)

Can be set to the individual integer pixel of N (number of the encoding block that N is predicted covering) by threshold value T here.If the value of InterToIntraFlag is 1, should carries out intraframe coding to inter-coded block A subsequently, otherwise inter-coded block A is carried out to interframe encode.

Owing in current video coding background ventrocephalad predictive frame, allowing to exist inter-coded macroblocks, so may have Intra-coded blocks in the encoding block that reference block A ' covers.In this case, if the area sum that predicted A ' of 8 encoding blocks that the area sum that referenced A ' of one or more Intra-coded blocks that reference block A ' covers covers and reference block A ' cover covers (, the area of reference block A ') ratio (for example is greater than certain threshold value, 3/4), in frame rate downsampling transcoding process subsequently, inter-coded block A is carried out to intraframe coding, otherwise inter-coded block A is carried out to interframe encode.

That is to say, the processing of obtaining the motion vector information reconfigured of inter-coded block A comprises: the ratio of the area that the original motion Vector Message of one or more encoding blocks that the reference block A ' according to the inter-coded block A in frame of video N in frame of video N-1 covers and the referenced A ' of each encoding block in the described one or more encoding blocks in frame of video N-1 cover and the area of reference block A ', obtain the approximate motion vector information of reference block; According to the approximate motion vector information of reference block and the original motion Vector Message of inter-coded block A, obtain the current motion vector information of inter-coded block A.

In the present invention, can there is the original motion Vector Message of inter-coded block of original motion Vector Message or the motion vector information reconfigured with inter-coded block of the motion vector information reconfigured by utilization, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured, thereby the inter-coded block that can in the down-sampling transcoding process, to neither having the original motion Vector Message, not have again a motion vector information reconfigured is carried out interframe encode.

Below with reference to specific embodiments of the invention, the present invention has been described, but those skilled in the art all understand, can carry out various modifications, combination and change to these specific embodiments, and can not break away from the spirit and scope of the present invention that limited by claims or its equivalent.

Can perform step with hardware or software as required.Note, without departing from the scope of the invention, can add step, therefrom remove step or revise step wherein to the flow chart provided in this specification.In general, flow chart just is used to refer to a kind of possible sequence for the basic operation of practical function.

Embodiments of the invention can utilize programming general purpose digital computer, utilize application-specific integrated circuit (ASIC), programmable logic device, field programmable gate array, light, chemistry, biological, system quantum or nanometer engineering, assembly and mechanism to realize.In general, function of the present invention can be realized by any means known in the art.Can use distributed or networked system, assembly and circuit.The communication of data or to transmit can be wired, wireless or by any other means.

Also will recognize, according to the needs of application-specific, one or more in the key element shown in accompanying drawing can be by more separating or more integrated mode realizes, or be removed even in some cases or be deactivated.Program or code that realization can be stored in machine readable media are carried out above-mentioned any method with the permission computer, also within the spirit and scope of the present invention.

In addition, it is only exemplary that any signal arrows in accompanying drawing should be considered to, rather than restrictive, unless concrete indication separately arranged.When term is also contemplated as, make to separate or the ability of combination while not knowing, the combination of assembly or step also will be considered to put down in writing.

Claims (18)

1. a frame rate downsampling transcoding method comprises:

Received compressing video frequency flow is carried out to partial decoding of h, and obtain the original coding information of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, described partial decoding of h and the video flowing after described partial decoding of h;

Current video frame in video flowing after described partial decoding of h is forward predicted frame and will be retained the time, for any one inter-coded macroblocks in described current video frame, in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block and/or the motion vector information reconfigured, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks, wherein, in described inter-coded macroblocks,

If the reference frame of an inter-coded block will be retained, this inter-coded block has former

The beginning motion vector information,

If the reference frame of an inter-coded block will be skipped, but this inter-coded block can reconfigure out at a described original motion Vector Message by one or more encoding blocks that on the frame of video be retained, the institute overlay area relates to by the reference block on its reference frame by this inter-coded block with respect to an approximate motion vector information by the frame of video that is retained, and this inter-coded block has the motion vector information reconfigured; And

Utilize the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after described partial decoding of h or the motion vector information reconfigured or reconstructed motion vector information and residual information, described partial decoding of h and the original coding information of each intra-coded frame in the video flowing after described partial decoding of h, the video flowing after described partial decoding of h is carried out to the frame rate downsampling transcoding.

2. frame rate downsampling transcoding method according to claim 1, is characterized in that, according to predetermined frame per second, judges that each frame of video in the video flowing after described partial decoding of h will be retained or will be skipped.

3. frame rate downsampling transcoding method according to claim 1, it is characterized in that, in the situation that there are one or more inter-coded block and one or more inter-coded block with the motion vector information reconfigured with original motion Vector Message in described inter-coded macroblocks simultaneously, utilize described one or more original motion Vector Message with inter-coded block of original motion Vector Message to obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks.

4. frame rate downsampling transcoding method according to claim 3, it is characterized in that, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, utilize described one of reference frame distance in described one or more inter-coded block with original motion Vector Message neither to there is the nearest original motion Vector Message with inter-coded block of original motion Vector Message of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither there is the original motion Vector Message but also do not there is the inter-coded block of the motion vector information reconfigured.

5. frame rate downsampling transcoding method according to claim 1, it is characterized in that, the reference block of an inter-coded block in described inter-coded macroblocks on its reference frame is in the situation that the related one or more encoding blocks in zone more than 3/4 by institute overlay area on the frame of video be retained have the original motion Vector Message, and a described inter-coded block is the inter-coded block with the motion vector information reconfigured.

6. frame rate downsampling transcoding method according to claim 1, it is characterized in that, in the situation that only there are one or more inter-coded block with original motion Vector Message in described inter-coded macroblocks, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, utilize described one of reference frame distance in described one or more inter-coded block with original motion Vector Message neither to there is the nearest original motion Vector Message with inter-coded block of original motion Vector Message of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither there is the original motion Vector Message but also do not there is the inter-coded block of the motion vector information reconfigured.

7. frame rate downsampling transcoding method according to claim 1, it is characterized in that, in the situation that only there are one or more inter-coded block with the motion vector information reconfigured in described inter-coded macroblocks, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, utilize described one of reference frame distance in described one or more inter-coded block with the motion vector information reconfigured neither to there is the nearest motion vector information reconfigured with inter-coded block of the motion vector information reconfigured of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither there is the original motion Vector Message but also do not there is the inter-coded block of the motion vector information reconfigured.

8. according to the described frame rate downsampling transcoding method of claim 4 or 6, it is characterized in that, obtain a described processing that neither there is the reconstructed motion vector information of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured and comprise:

Select before a described reference frame that neither there is the inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured, approach a described reference frame that neither has an inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured most one using the frame of video that is retained as described one new reference frame that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured;

Obtain described one according to following equation and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured:

${\mathrm{MV}}_{B*-B}={\mathrm{MV}}_{A^{\′}-A}*\frac{N-(N-n-r)}{N-(N-m)}$

Wherein, N means the frame number of described current video frame, N-m means that nearest one of the described reference frame that neither has an inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance has the frame number of reference frame of the inter-coded block of original motion Vector Message, N-n-r means a described frame number that neither has the new reference frame of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured

mean that described one neither has the original motion Vector Message and do not have again the reconstructed motion vector of the inter-coded block of the motion vector information reconfigured, MV _{a '-A}mean the described nearest original motion vector with inter-coded block of original motion Vector Message of reference frame that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance.

9. according to the described frame rate downsampling transcoding method of claim 5 or 7, it is characterized in that, obtain a described processing that neither there is the reconstructed motion vector information of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured and comprise:

Select before a described reference frame that neither there is the inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured, approach a described reference frame that neither has an inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured most one using the frame of video that is retained as described one new reference frame that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured;

Obtain described one according to following equation and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured:

${\mathrm{MV}}_{B*-B}={\mathrm{MV}}_{A^{\′}-A}*\frac{N-(N-n-r)}{N-(N-m)}$

Wherein, N means the frame number of described current video frame, one of meaning that nearest one of the described reference frame that neither has an inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance has that the reference block of the inter-coded block of the motion vector information reconfigured covers of N-m is by the frame number of the frame of video that is retained, N-n-r means a described frame number that neither has the new reference frame of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured mean that described one neither has the original motion Vector Message and do not have again the reconstructed motion vector of the inter-coded block of the motion vector information reconfigured,

mean the described nearest motion vector reconfigured with inter-coded block of the motion vector information reconfigured of reference frame that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance.

10. a frame rate downsampling transcoding device comprises:

The partial decoding of h unit, be configured to received compressing video frequency flow is carried out to partial decoding of h, and obtain the original coding information of each intra-coded frame comprised in the original coding information of each Intra-coded blocks in each forward predicted frame comprised in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame comprised in the video flowing after partial decoding of h and residual information, described partial decoding of h and the video flowing after described partial decoding of h;

The vector reconstruction unit, be configured to current video frame in the video flowing after described partial decoding of h and be forward predicted frame and will be retained the time, for any one inter-coded macroblocks in described current video frame, in the situation that the motion vector information that the one or more inter-coded block in described inter-coded macroblocks have the original motion Vector Message and/or reconfigure, utilize the original motion Vector Message of described one or more inter-coded block and/or the motion vector information reconfigured, obtain and neither there is the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks, wherein, in described inter-coded macroblocks,

If the reference frame of an inter-coded block will be retained, this inter-coded block has the original motion Vector Message,

If the reference frame of an inter-coded block will be skipped, but this inter-coded block can reconfigure out at a described original motion Vector Message by one or more encoding blocks that on the frame of video be retained, the institute overlay area relates to by the reference block on its reference frame by this inter-coded block with respect to an approximate motion vector information by the frame of video that is retained, and this inter-coded block has the motion vector information reconfigured; And

The down-sampling transcoding units, be configured to utilize the original coding information of each Intra-coded blocks in each forward predicted frame in the video flowing after the original motion Vector Message of each inter-coded block in each forward predicted frame in the video flowing after described partial decoding of h or the motion vector information reconfigured or reconstructed motion vector information and residual information, described partial decoding of h and the original coding information of each intra-coded frame in the video flowing after described partial decoding of h, the video flowing after described partial decoding of h is carried out to the frame rate downsampling transcoding.

11. frame rate downsampling transcoding device according to claim 10, is characterized in that, described frame rate downsampling transcoding device judges that according to predetermined frame per second each frame of video in the video flowing after described partial decoding of h will be retained or will be skipped.

12. frame rate downsampling transcoding device according to claim 10, it is characterized in that, in the situation that there are one or more inter-coded block and one or more a plurality of inter-coded block with the motion vector information reconfigured with original motion Vector Message in described inter-coded macroblocks simultaneously, described vector reconstruction unit utilizes described one or more original motion Vector Message with inter-coded block of original motion Vector Message to obtain neither to have the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured in described inter-coded macroblocks.

13. frame rate downsampling transcoding device according to claim 12, it is characterized in that, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, described vector reconstruction unit utilizes described one of the reference frame distance in described one or more inter-coded block with original motion Vector Message neither to have the nearest original motion Vector Message with inter-coded block of original motion Vector Message of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured.

14. frame rate downsampling transcoding device according to claim 10, it is characterized in that, the reference block of an inter-coded block in described inter-coded macroblocks on its reference frame is in the situation that the related one or more encoding blocks in zone more than 3/4 by institute overlay area on the frame of video be retained have the original motion Vector Message, and a described inter-coded block is the inter-coded block with the motion vector information reconfigured.

15. frame rate downsampling transcoding device according to claim 10, it is characterized in that, in the situation that only there are one or more inter-coded block with original motion Vector Message in described inter-coded macroblocks, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, described vector reconstruction unit utilizes described one of the reference frame distance in described one or more inter-coded block with original motion Vector Message neither to have the nearest original motion Vector Message with inter-coded block of original motion Vector Message of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured.

16. frame rate downsampling transcoding device according to claim 10, it is characterized in that, in the situation that only there are one or more inter-coded block with the motion vector information reconfigured in described inter-coded macroblocks, neither there is the original motion Vector Message for one and do not there is again the inter-coded block of the motion vector information reconfigured, described vector reconstruction unit utilizes described one of the reference frame distance in described one or more inter-coded block with the motion vector information reconfigured neither to have the nearest motion vector information reconfigured with inter-coded block of the motion vector information reconfigured of reference frame that the original motion Vector Message do not have not only an inter-coded block of the motion vector information reconfigured and obtain a described reconstructed motion vector information that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured.

17. according to the described frame rate downsampling transcoding device of claim 13 or 15, it is characterized in that, described vector reconstruction unit obtains described one by following processing and neither has the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured:

Select before a described reference frame that neither there is the inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured, approach a described reference frame that neither has an inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured most one using the frame of video that is retained as described one new reference frame that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured;

Obtain motion vector information before a described reconstruction that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured according to following equation:

${\mathrm{MV}}_{B*-B}={\mathrm{MV}}_{A^{\′}-A}*\frac{N-(N-n-r)}{N-(N-m)}$

Wherein, N means the frame number of described current video frame, N-m means that nearest one of the described reference frame that neither has an inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance has the frame number of reference frame of the inter-coded block of original motion Vector Message, N-n-r means a described frame number that neither has the new reference frame of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured

mean that described one neither has the original motion Vector Message and do not have again the reconstructed motion vector of the inter-coded block of the motion vector information reconfigured, MV _{a '-A}mean the described nearest original motion vector with inter-coded block of original motion Vector Message of reference frame that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance.

18. according to the described frame rate downsampling transcoding device of claim 14 or 16, it is characterized in that, described vector reconstruction unit obtains described one by following processing and neither has the reconstructed motion vector information that the original motion Vector Message does not have again the inter-coded block of the motion vector information reconfigured:

Select before a described reference frame that neither there is the inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured, approach a described reference frame that neither has an inter-coded block that the original motion Vector Message do not have not only the motion vector information reconfigured most one using the frame of video that is retained as described one new reference frame that neither has the original motion Vector Message but also do not have the inter-coded block of the motion vector information reconfigured;

Obtain motion vector information before a described reconstruction that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured according to following equation:

${\mathrm{MV}}_{B*-B}={\mathrm{MV}}_{A^{\′}-A}*\frac{N-(N-n-r)}{N-(N-m)}$

Wherein, N means the frame number of described current video frame, one of meaning that nearest one of the described reference frame that neither has an inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance has that the reference block of the inter-coded block of the motion vector information reconfigured covers of N-m is by the frame number of the frame of video that is retained, N-n-r means a described frame number that neither has the new reference frame of the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured mean that described one neither has the original motion Vector Message and do not have again the reconstructed motion vector of the inter-coded block of the motion vector information reconfigured,

mean the described nearest motion vector reconfigured with inter-coded block of the motion vector information reconfigured of reference frame that neither there is the inter-coded block that the original motion Vector Message do not have again the motion vector information reconfigured of described reference frame distance.

Images