CN1192631C - Method and device for coding and decoding digitized frames of special effects film - Google Patents
Method and device for coding and decoding digitized frames of special effects film Download PDFInfo
- Publication number
- CN1192631C CN1192631C CNB971154252A CN97115425A CN1192631C CN 1192631 C CN1192631 C CN 1192631C CN B971154252 A CNB971154252 A CN B971154252A CN 97115425 A CN97115425 A CN 97115425A CN 1192631 C CN1192631 C CN 1192631C
- Authority
- CN
- China
- Prior art keywords
- frame
- film
- film scene
- scene
- prediction
- 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.)
- Expired - Lifetime
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/004—Predictors, e.g. intraframe, interframe coding
Abstract
A method for coding digitized frames of a stroboscope-like special effects film is proposed, in which alternately a first number of frames of a first film scene and a second number of frames of a second film scene is shown. Two frame buffers (30, 32) are available for coding the frames. The method is characterized by the fact that the two frame buffers (30, 32) are operated separately so that in each case one of the frame buffers (30) is permanently used for predicting frames of the first film scene and the other frame buffer (32) is permanently used for predicting frames of the second film scene. In addition, a corresponding method for decoding frames coded in this manner and devices for coding and decoding digitized frames of a stroboscope-like special effects film are also proposed.
Description
The present invention is relevant to method and the device that the digital frame of special effects film is encoded and deciphered.
The present invention is based on a kind of method that the digital frame of the special effects film of general type is encoded and deciphered.
Commonly known in the art common with relative options qualification seldom to full film image Methods for Coding.The MPEG2 standard has begun to change this situation, because it has allowed the optional structure of larger amt comparatively speaking.
For the following standard of the full film image of encoding, require them that greater flexibility should be arranged, make coding parameter can adapt to the feature of the image document that will be encoded on than bigger up to now degree.In this connection, can consider the special parameter of vision signal class, this does not especially frequently occur, but can be encoded better by the special selection to parameter yet.
Such class vision signal is from the frame of the conversion multiterminal of two frame sequences alternately, and this frame sequence frequently takes place when music image clips for example.In this special effects film, first some frames back of first film scene follows hard on second some frames of second film scene.Therefore, provide the frequent scene of a complete new images content to edit out in present this special effects film in each case.A kind of like this special effects film is digitized and subsequently according to the MPEG2 standard code of being familiar with, frequent scene editor causes lot of data at the output of code device, this is because because to the motion prediction of the success of each macro block of first frame, code device can not be carried out any redundant compression at this frame behind the scene editor, thereby produced a large amount of so-called interior coded macroblockss, they do not have temporarily to predict and encode at first frame.Compare with unidirectional or bi-directional predictive coding piece, the data volume of these macro blocks is much bigger.Under the situation of unidirectional or bi-directional predictive coding piece, only the macro block that will be transmitted and the difference between the motion compensation macro block are encoded.
The objective of the invention is to propose a kind of to described special effects film Methods for Coding, although frequent scene editor is wherein taken place, but data volume is still reduced greatly, for this reason, in the coding method such, only require less change, and in fact hardware complexity remains unchanged fully such as MPEG2.
According to the present invention, two frame buffers in the code device are controlled respectively, the mode of operation of these two frame buffers, make in each case, a frame buffer is used to predict the frame of first film scene, and another frame buffer is used to predict all frames of second film scene, if do not predict, two frame buffers all are used to one of them a frame of bi-directional predicted two film scenes so.Compare with method commonly known in the art, the advantage of this method is when scene changes, and the last frame of the film scene of transferring still is stored in one of them frame buffer.As a result, can be in frame buffer the new frame that will encode be carried out motion estimation successively on the basis of last frame.Thereby even after the scene editor, interior coded macroblocks also needn't be made existing, and data volume is significantly reduced.
If have only so-called forward prediction be used to the to encode frame of special effects film, so can advantageous particularly.In forward prediction, the motion estimation of the frame that be encoded is carried out with a reference frame, from time sequencing, this reference frame appears at the frame that will be encoded in the frame sequence that will be presented at subsequently on the screen before.By being the forward prediction of predictive mode regulation, this coding method is greatly simplified, and reason is not need frame is classified again in cataloged procedure.
If forward and backward prediction are all licensed as predictive mode, if frame is classified so, make the frame of backward prediction be delayed coding again, and before the frame of backward prediction, be cooled by its frame that begins to carry out backward prediction, can be favourable.This classification more also is provided in the MPEG2 standard.Under the situation of backward prediction frame, carry out on the basis to the frame of motion compensation after only being presented at the frame that will be encoded of the frame that will be encoded by the time order.
If regulation is inserted in the data flow of each predictive frame by the information that it begins to carry out the frame buffer of frame prediction, also is favourable.This information belongs to the information of which film scene corresponding to the regulation frame.Thus, a code translator can learn immediately which frame buffer will be used to motion compensation.
Being inserted in the data flow of each predictive frame if stipulate the information that this frame belongs to the special effects film sequence of conversion multiterminal, also is favourable.By this information, code translator can be reconstructed, and makes that for example one only allows forward prediction or allows the pattern of forward and backward prediction to be preestablished.
Correspondingly, concerning the method for the frame coding of the method according to this invention coding, if two frames in the code translator are operated respectively, make same frame buffer always be used to predict the frame of first film scene, and another frame buffer is used to predict the frame of second film scene.Even this makes after scene change, the frame of the motion compensation that code device is also right coding, thus the data volume of the frame that is encoded is reduced.
In order to simplify the image processing in the code translator, also can be favourable if only allow forward prediction to come the special effects film of predictive transformation multiterminal.
On the other hand, if the order that is output corresponding to frame is classified (being like this to the backward prediction frame at least) according to the MPEG2 standard again, so also can allow forward prediction or backward prediction.
If the information that has been inserted in the data flow of predictive frame is used to determine the frame buffer that must be begun predictive frame by it, also is favourable.This has eliminated the needs of research unit of complexity of determining to be begun by it frame buffer of predictive frame to independent.
There is shown three exemplary embodiments of the present invention, and in the following description it has been done detailed explanation.In the accompanying drawing:
Fig. 1 shows the series of frames of the special effects film sequence of conversion multiterminal, wherein has only I and P frame to be encoded, and has only adopted forward prediction;
Fig. 2 shows the series of frames of the special effects film sequence of conversion multiterminal, and wherein I, B and P frame are used, and have only forward prediction to be allowed to;
Fig. 3 shows the series of frames of the special effects film sequence of conversion multiterminal, and wherein I, B and P frame are used, and forward prediction and backward prediction all are allowed to;
Fig. 4 shows the block diagram to a code device of the frame coding of the special effects film of conversion multiterminal;
Fig. 5 shows the block diagram to a code translator of the frame coding of the special effects film of conversion multiterminal;
Fig. 6 illustrates the data flow of a coded frame;
Fig. 7 shows a table of the information that comprises relevant coded frame, this information be inserted in coded frame data flow begin the place.
Fig. 1 shows an example of the special effects film sequence of conversion multiterminal.In this sequence, four frames of second film scene are followed four frames of first film scene in each case, and these frames are provided with subscript in each case.The corresponding frame of subscript 1 expression belongs to first film scene, and the corresponding frame of subscript 2 expressions belongs to second film scene.First frame that illustrates is the I frame, and this frame is corresponding to an intra coded frame that defines in the MPEG2 standard.In this respect, can be clearly with reference to the video section of MPEG2 standard, this standard is numbered ISO/IEC 13818-2 in iso standard.
When taking place, the coding of this I frame do not do motion compensation by means of another frame.After this first I frame is three P frames of first film scene.This P frame promptly is used as a frame that more lags behind in time of motion compensation reference corresponding to the frame with motion compensation encoding.Also comprise this qualification in the MPEG2 standard to the P frame.Arrow on the right of each P frame sensing of Fig. 1 indicates corresponding P frame to do motion compensation by means of forward prediction.The preceding four frame back of first film scene are preceding four frames of second film scene.First frame of second film scene is again an I frame, and three frames of ensuing second film scene are the P frames of having done forward prediction equally.After four frames of second film scene is four P frames of first film scene.The 5th frame of first film scene is a P frame now, and it has done forward prediction from the 4th frame of first film scene.Second four frame back of first film scene is again four frames of second film scene.The 5th frame of second film scene also is a P frame, and is encoded by means of forward prediction from the 4th frame of second film scene.Other frame sequence of special effects film sequence does not further illustrate.Yet it is corresponding continuation, but in each case, can insert an I frame again to two film scenes at interval at some, thereby picture quality can not reduced by a plurality of forward predictions.Delegation below the frame shown in the figure shows the coded sequence of frame.Can clearly be seen that, frame is done classification again, but each frame is the same order coding that appears at the input of code device with it.Below the frame sequence shown in the figure second row has been specified frame buffer, wherein imports corresponding frame.Symbol * indication associated frame is written to first frame buffer of a code device, and character # indication associated frame is written to second frame buffer of a code device.Should be noted that frame buffer has only the space of using for a frame in each case, therefore, when writing a frame again in one frame buffer, previously stored frame is rewritten.Fig. 1 shows when for example encoding the 5th frame of first film scene, and the 4th frame of first film scene still is stored in the relevant frame buffer, thereby can carry out forward prediction on the 4th frame basis of first film scene.Otherwise also as can be seen, for example when the 5th frame of coding second film scene, the 4th frame of second film scene still is stored in the associated frame buffer, thereby can carry out forward prediction on the 4th frame basis of second film scene.
Fig. 2 shows another frame sequence of the special effects film sequence of conversion multiterminal.Equally, four frames of second film scene are followed four frame back at first film scene, but at this moment except I frame and P frame, also to so-called B frame coding.According to the MPEG2 standard, the B frame is corresponding to a frame of passive movement compensation coding, the frame after promptly leaning in time and/or be used as following closely the frame in time of motion compensation reference.Concerning coding method shown in Figure 2, shown in arrow, B frame means that it is only predicted by the frame after leaning on the time, that is to say and has adopted forward prediction.From this respect, it is corresponding to the P frame, but has following difference at least with the latter, promptly when this B frame of coding, allows bigger predicated error tolerance limit.Therefore, a B frame produces littler data volume than P and I frame.It is predicted on the 4th frame basis of first film scene that Fig. 2 illustrates the 5th, the 6th, the 7th frame of first film scene for example.Yet this only just might not realize when the 5th of first film scene is imported into relevant frame buffer with the 6th frame.This as can be seen situation of second row below frame sequence, the coding order that therefrom can see frame in this exemplary embodiment are also corresponding to the chronological order of the frame of code device input.
Fig. 3 shows the other frame sequence of the special effects film of conversion multiterminal.I, B and P frame have also appearred in this sequence.Yet with the frame comparison of Fig. 2, difference is that the B frame no longer only carries out motion compensation by forward prediction, and in this case, forward prediction and backward prediction can be adopted.Arrow from Fig. 3 as can be seen, some B frame of first film scene is with backward prediction compensation, the front cross frame of second film scene, the 6th frame and the 7th frame also carry out motion compensation by backward prediction.Yet the frame of backward prediction must be classified according to the coding order subsequently again.Shown in the delegation below the frame sequence among Fig. 3, after first frame of first film scene was encoded, the 4th frame of first film scene at first was encoded, and then the second and the 3rd frame is encoded.Under this arrangement, the backward prediction of the second and the 3rd frame of first film scene is to carry out on the 4th frame basis of first film scene.Next the frame that will be encoded is the 3rd frame of second film scene, and this is an I frame.After this first frame and second frame of second film scene are encoded.To the 6th frame coding of first film scene, it is encoded as the P frame subsequently.After this could be according to the 3rd frame of second film scene by encode the 4th frame of second film scene of forward prediction, subsequently by the 5th frame coding of backward prediction to first film scene.Second row with reference to figure 3 can be easy to find out other sequence.Equally, can find out to have only the I frame to be imported into relevant frame buffer at an easy rate from the third line of Fig. 3 with the P frame.The change of coding order is corresponding to the coding order of in the MPEG2 standard such I, a B, P frame sequence being advised.
Next will be with reference to the structure of figure 4 descriptions according to code device of the present invention.The video input of label 20 indication code devices provides digitized frame at this input.Taxon of label 21 indications is carried out above-mentioned classification again to frame in this unit, so that change the coding order under every kind of situation, make and encode after the B frame is than P frame in each case, or the I frame is after the B frame.Yet taxon 21 only is provided as a selectable unit, and it can be omitted in the embodiment of Fig. 1 and Fig. 2 fully.Subtrator of label 22 indications, it is connected to a DCT unit, carries out discrete cosine transform in this DCT unit, and this also is used in the code device of prior art.Quantifying unit of label 24 indications, it is with the coefficient quantization that obtains in the DCT unit 23.VLC unit of label 25 indications in this unit, carries out variable length code to coefficient after dct transform.In addition, this unit also plays the function of multiplexer, so that other information also can be inserted in the data flow at this point.
The data flow that produces is sent to buffer storage 27, provides the frame that is encoded by the coding order at the output of buffer storage 27.Quantifying unit is gone in label 28 one of indication, thereby is being gone quantifying unit to do the re-quantization processing immediately from the data of quantifying unit 24.After this, carry out the inverse transformation of discrete cosine transform at an IDCT unit, thereby the frame of reconstruct appears in the output in IDCT unit 29.
This frame is via switch element 35 or be input to first mc unit 31, perhaps is input to second mc unit 33.Mc unit 31,33 is carried out motion compensation to the frame of reconstruct respectively.For this reason, it adopts a reference frame that still is stored in respectively in the first and second relevant frame buffers 30,32.Mc unit determines to be sent to via MV bus 36 motion vector of VLC unit 25.Mc unit 31 and 33 all conducts interviews to MV bus 36.Movement compensating frame is presented on the output of mc unit 31 or 33 respectively subsequently, and it offers subtrator 22 via second switch unit 34.For simplicity, this output illustrates as the output from the frame buffer among Fig. 4 30,32 in each case.The frame that produces inputs 20 at subtrator 22 is poor with relevant movement compensating frame, poor between 23 to 25 processing in unit, two frames then, thus reduce data volume significantly.Also to should be mentioned that, produce difference and carry out motion compensation only to B frame and the execution of P frame, the I frame is not done motion compensation via unit 30 to 35 at subtrator 22.
Next will be via 38, one external unit indications of special effects film identification signal input to a special effects film coding, wherein the frame of second quantity of the frame of first quantity of first film scene and second film scene is alternately encoded continuously.External unit can be the part of studio apparatus.This external unit also via 37 while of input dump movie scene Recognition signal, is used for notifying code device the current frame of first film scene or the frame of second film scene of whether existing subsequently.This signal is used to control switch unit 34 and 35 subsequently, and this has guaranteed that prediction always takes place from correct frame buffer 30,32, depends on that it still is second film scene that the frame that will encode belongs to first film scene.When cataloged procedure finished, the reconstructed frame that occurs at the output of IDCT unit was imported into relevant frame buffer 30,32, but this carries out I frame and P frame, and the B frame was not carried out.
What it might also be mentioned is that whole code device can dispose via special effects film identification signal input 38.For example, in embodiment illustrated in figures 1 and 2, mc unit 31 and 33 can be set to only allow forward prediction.And in the embodiment shown in fig. 3, mc unit can be set to the B frame is allowed forward and backward prediction simultaneously.Corresponding information in input 37 and 38 also is transmitted to VLC unit 25, and this unit is inserted into the adequate information at a suitable some place in the data flow of a coded frame, will be described with reference to figure 6 and Fig. 7 below this.
Below with reference to the structure of Fig. 5 description according to a code translator of the present invention.In Fig. 5, put on the label identical with Fig. 4 components identical, so these elements will be described in detail no longer with Fig. 4.The frame that is encoded is presented on the input 49 of code translator according to its coding order, they at first are sent to buffer storage 27, be a VLD unit 50 afterwards, this unit is used to decipher the variable length code according to VLC unit 25 codings of Fig. 4, it also is used as demultiplex unit, be used for selecting some information of data flow, below will be explained in detail this.Decoding data is provided for quantifying unit 28.29 inverse transformations of carrying out discrete cosine transform in the IDCT unit, and the data that produce are sent to adder unit 51.At last, the frame of generation quilt is with respect to going the frame order in classification (desorting) unit 52 to classify again.These frames are not with the coding order, but the frame sequence output that presents at the input 20 of code device with them.
With regard to the motion compensation decoding of frame, first frame buffer 30, second frame buffer 32, first mc unit 31 and second mc unit 33 are also provided.The data of motion vector can be via 36 inputs of MV bus in mc unit 31 and 33.The frame that occurs at the output of adder unit 51 is written to one of them of frame buffer 30 or 32 respectively, as the reference frame of finishing decoded operation.Here should be noted that to have only I frame and P frame to be written into one of them of frame buffer, and do not have the B frame.Frame buffer is controlled corresponding to the information in the data flow of film scene identification signal and special effects film identification signal by means of inserting with relevant mc unit.If do not receive special effects film identification signal information, control according to the MPEG2 standard so.
This information leaches from data flow in VLD unit 50.Information corresponding to the film scene identification signal is sent to switch element 53 and 54 via circuit 55 then.In addition, actually or about wanting decoded frame I frame, the information of P frame B frame also to work at switch element.If B frame, switch element 54 is set to its 3rd position so, at this moment the frame of the output of adder unit 51 is not imported into any one in two frame buffers 30 or 32, and switch element 53 causes the corresponding film scene that is provided for adder unit 51 is in each case done motion compensation from the associated frame buffer to frame.To will be by the frame of VLD unit 50 decoding, also from data flow, leach and the corresponding information of special effects film identification signal, this information is sent to mc unit 31 and 33 via circuit 56, it provides the suitable reconstruct to these unit, so that macro block to the frame that for example all will be decoded, forward prediction is carried out in permission in proper order, perhaps as an alternative, the macro block of B frame is allowed forward and backward prediction simultaneously.
Fig. 6 shows the rough structure of the data flow of a coded frame.Reference symbol PH indication comprises a data field about the general information of coded frame.Reference symbol SLH indication comprises the data field of relevant information of a horizontal line of the macro block of the frame that will be encoded.Reference symbol MB1, MB2 or the like indicate the data field of the related frame data of the respective macroblock that comprises this horizontal line.One frame comprises the horizontal line of relatively large quantity, as shown in Figure 6.The starting point of next frame also is shown in Fig. 6.
The other information that is used for controlling the frame buffer of code translator is inserted into the data field PH of the information of the frame of wanting decoded in addition.
Fig. 7 shows the structure according to the data field PH of MPEG2 standard, and according to the present invention, this structure has been done expansion by extra information.Each information by numeral 1 to 23 indication has been done definition in the MPEG2 standard, therefore do not need detailed explanation at this.According to the present invention, fresh information 24 and 25 is inserted in the data field in addition.Information " stroboscope_mode " is used to indicate this frame to belong to the special effects film sequence of conversion multiterminal.This information is added one just enough.Label is that another information of 25 (scene 1 sign) represents that it still is second film scene that this frame belongs to first film scene actually, and this information only just exists when " stroboscope_mode " information is inserted in the data flow.Because other information does not define in the MPEG2 standard, the code translator that therefore only is in particular this purpose design could correctly be estimated this information.As another kind of scheme, two items of information also can be inserted in the data flow more in addition, thereby as the expansion of the data field of general frame information.This extra expansion allows in the MPEG2 standard.
Claims (19)
1. the digital frame of the video film of second number of frames of first number of frames that alternately provides first film scene and second film scene is carried out the method for encoder motion prediction, utilize two frame buffers (30,32) these frames of encoding, this method comprises the following steps:
Response is notified the frame of the current frame that whether has first film scene or second film scene and is used to control the control signal of choice device, belong at described frame under the situation of first film scene, select a frame buffer (30) to come described frame is carried out motion prediction by choice device (34,35);
Respond described control signal, belong at described frame under the situation of second film scene, select another frame buffer (32) to come described frame is carried out motion prediction by choice device (34,35);
Based on described motion prediction step, described frame is carried out redundant compressed encoding,
Wherein said control signal provides from outside film scene Recognition signal generating unit.
2. method according to claim 1 wherein only adopts forward prediction to predict the frame of first or second film scene.
3. method according to claim 1, wherein adopt the frame of predicting first or second film scene forward or backwards, at least for the frame of backward prediction, according to the MPEG2 standard coding order is classified again, so that the frame of time-delay coding backward prediction, and the reference frame that coding carries out backward prediction according to this before the frame of this backward prediction.
4. according to one of them described method of foregoing claim, wherein information is inserted in the data flow of each predictive frame at least, go out the frame buffer (30,32) that need be used to predict by this information inference.
5. method according to claim 1, can be used for wherein determining that the information of special effects film sequence of frame that whether this frame belongs to second quantity of the frame of first quantity that first film scene alternately is shown therein and second film scene inserts in the data flow of each predictive frame at least, thereby frame buffer (30,32) must separate operation.
6. the method that the frame of the film of the frame of second quantity of the frame of first quantity that alternately provides first film scene of method according to claim 1 coding and second film scene is deciphered, two frame buffers (30 are provided, 32) decipher these frames, this method comprises the following steps:
Carry out the inverse operation of the redundant compressed encoding of frame;
Response is notified the frame of the current frame that whether has first film scene or second film scene and is used to control the control signal of choice device, belonging under the situation of first film scene through the frame after the described inverse operation step process, by choice device (34,35) select a frame buffer (30) to come described frame is carried out motion prediction, described control signal is to receive with the frame data of encoding;
Respond described control signal,, select another frame buffer to come described frame is carried out motion prediction by choice device (34,35) belonging under the situation of second film scene through the frame after the described inverse operation step process.
7. method according to claim 6 wherein only adopts forward prediction to predict the frame of first or second film scene.
8. method according to claim 6, wherein adopt the frame of predicting first or second film scene forward or backwards, at least to the frame of backward prediction, according to the MPEG2 standard output order is classified again, so that export the frame of backward prediction in time in advance, and the reference frame that backward prediction is carried out in output according to this after the frame of this backward prediction.
9. according to the described method of one of claim 6 to 8, wherein be inserted into a item of information in the data flow of each predictive frame at least and be used to determine and be used for the frame buffer of predictive frame.
10. the digital frame of the film of the frame of second quantity of the frame of first quantity of alternately following first film scene and second film scene is carried out the equipment of encoder motion prediction, have two frame buffers (30,32), described equipment is characterised in that and comprises:
Choice device (34,35), response is notified the frame of the current frame that whether has first film scene or second film scene and is used to control the control signal of choice device, belong at described frame under the situation of first film scene, select a frame buffer (30) to predict described frame, and belong at described frame under the situation of second film scene, select another frame buffer (32) to predict described frame, wherein said control signal provides from outside film scene Recognition signal generating unit;
Redundant compression-encoding device (22,23,24,25) is used for based on described prediction described frame being carried out redundant compressed encoding.
11. equipment according to claim 10, first and second switch elements (34 wherein are provided, 35) as selecting frame buffer (30,32) one of choice device, they are controlled by film scene marking signal (37), predict the frame that will be encoded to select first or second frame buffer (30,32).
12. according to claim 10 or 11 described equipment, wherein provide data to insert device (25), this device is according to the film scene marking signal (37) that is input to this equipment, information is inserted the data flow of each predicted frame at least, this information is specified the frame buffer (30,32) that need be used to carry out the frame prediction.
13. equipment according to claim 12, wherein data are inserted device (25) and are being detected special effects film marking signal (38) afterwards, the information of inserting in addition is in the data flow of each frame of special effects film, this information indicates this frame to belong to a special effects film sequence, and the frame of second quantity of the frame of first quantity of first film scene and second film scene is alternately followed each other in this sequence.
14. equipment according to claim 13 wherein provides the device that predictive mode is set, this device for good and all is set to predictive mode with the forward prediction pattern after detecting special effects film marking signal (38) existence.
15. equipment according to claim 13, the device that predictive mode is set wherein is provided, this device is after detecting special effects film marking signal (38) existence, with forward and backward prediction all as predictive mode, at least to the frame of backward prediction, according to the MPEG2 standard coding order is classified again, so that the frame of time-delay coding backward prediction, and the reference frame that coding carries out backward prediction according to this before the frame of this backward prediction.
16. to the equipment of deciphering according to the frame of the film of the frame of second quantity of the frame of first quantity of alternately following first film scene of the method for claim 1 coding and second film scene, have two frame buffers (30,32), described equipment is characterised in that and comprises:
The redundant compressed encoding of frame is carried out the device (50,28,29,51) of inverse operation;
Choice device (54,53), response is notified the frame of the current frame that whether has first film scene or second film scene and is used to control the control signal of choice device, belong under the situation of first film scene at the frame that provides from the described output that carries out the device of inverse operation, select a frame buffer (30) to predict described frame, and belong under the situation of second film scene at the frame that provides from the described output that carries out the device of inverse operation, select another frame buffer (32) to predict the frame of second film scene, described control signal is to receive with the frame data of encoding.
17. equipment according to claim 16, first and second switch elements (53 wherein are provided, 54) as selecting one of frame buffer or another choice device, they by in the data flow of wanting decoded frame corresponding to the item of information control of the frame buffer that is used to predict (30,32).
18. according to claim 16 or 17 described equipment, the device that predictive mode is set wherein is provided, after the item of information in the data flow that detects the frame of wanting decoded, the sign that belongs to the special effects film sequence corresponding to frame, the frame of second quantity of the frame of first quantity of first film scene and second film scene is alternately followed each other in this sequence, and this device for good and all is set to predictive mode with the forward prediction pattern.
19. according to claim 16 or 17 described equipment, the device that predictive mode is set wherein is provided, after notification frame belongs to the item of information of special effects film sequence in the data flow that detects the frame of wanting decoded, the frame of second quantity of the frame of first quantity of first film scene and second film scene is alternately followed each other in this sequence, this device with forward and backward prediction all as predictive mode, at least to the frame of backward prediction, according to the MPEG2 standard output order is classified again, so that export the frame of backward prediction in time in advance, and the frame that backward prediction is carried out in output according to this after the frame of this backward prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB971154252A CN1192631C (en) | 1996-07-26 | 1997-07-23 | Method and device for coding and decoding digitized frames of special effects film |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DE19630295A DE19630295A1 (en) | 1996-07-26 | 1996-07-26 | Method for coding and decoding digitized pictures of an animated film and device for coding and decoding digitized pictures of an animated film |
| DE19630295.1 | 1996-07-26 | ||
| CNB971154252A CN1192631C (en) | 1996-07-26 | 1997-07-23 | Method and device for coding and decoding digitized frames of special effects film |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1179669A CN1179669A (en) | 1998-04-22 |
| CN1192631C true CN1192631C (en) | 2005-03-09 |
Family
ID=34634855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB971154252A Expired - Lifetime CN1192631C (en) | 1996-07-26 | 1997-07-23 | Method and device for coding and decoding digitized frames of special effects film |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1192631C (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6819715B2 (en) * | 2000-07-25 | 2004-11-16 | Matsushita Electric Industrial Co., Ltd. | Image decoding method, image decoding apparatus, and data storage medium |
| KR100498363B1 (en) * | 2003-08-07 | 2005-07-01 | 엘지전자 주식회사 | Method for post-processing decoded image of mobile phone |
-
1997
- 1997-07-23 CN CNB971154252A patent/CN1192631C/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| CN1179669A (en) | 1998-04-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1134172C (en) | Emboding supplemental data in digital video signal | |
| US6058140A (en) | Method and apparatus for inverse 3:2 pulldown detection using motion estimation information | |
| EP0542195B1 (en) | Interframe predicting method and picture signal coding/decoding apparatus | |
| CN1098599C (en) | System for coding video signal in presence of image intensity gradient | |
| CN101969566B (en) | Moving image encoding device and moving image encoding method | |
| CN101690234B (en) | Video encoding method, decoding method, device thereof | |
| US5815217A (en) | Scene-change detector for a moving image composed of a plurality of frames | |
| CN101990761B (en) | Moving image encoding apparatus and moving image encoding method | |
| CN1201598C (en) | Encoding of two correlated sequences of data | |
| CN101005614A (en) | Coding and decoding method and device for dynamic image, and computer readable record medium | |
| CN1115548A (en) | A method and device for estimating movement | |
| CN1157953C (en) | Detection of watermark in compressed video signal | |
| CN102196253B (en) | Video coding method and device based on frame type self-adaption selection | |
| JP4518599B2 (en) | 3: 2 pulldown detection and optimized video compression encoder in motion estimation phase | |
| WO1995004432A1 (en) | Coding image data | |
| CN1192631C (en) | Method and device for coding and decoding digitized frames of special effects film | |
| US6292199B1 (en) | Method and device for copying and decoding digitized frames of a special effects film | |
| US6473465B1 (en) | Method and apparatus for video coding at high efficiency | |
| US6556714B2 (en) | Signal processing apparatus and method | |
| KR19980033415A (en) | Apparatus and method for coding / encoding moving images and storage media for storing moving images | |
| CN1516851A (en) | Method and device for detecting watermark | |
| US7397856B2 (en) | Method and apparatus for storing and processing picture information of temporally successive pictures | |
| CN1748427A (en) | Predictive encoding of motion vectors including a flag notifying the presence of coded residual motion vector data | |
| US20070183673A1 (en) | Video processing method and corresponding encoding device | |
| Dorai et al. | Extracting motion annotations from MPEG-2 compressed video for HDTV content management applications |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: GR Ref document number: 1034422 Country of ref document: HK |
|
| CX01 | Expiry of patent term |
Granted publication date: 20050309 |
|
| CX01 | Expiry of patent term |