CN105306950A - Video compression remote transmission system for feeding back coarse quantization reconstructed frames - Google Patents
Video compression remote transmission system for feeding back coarse quantization reconstructed frames Download PDFInfo
- Publication number
- CN105306950A CN105306950A CN201510889210.5A CN201510889210A CN105306950A CN 105306950 A CN105306950 A CN 105306950A CN 201510889210 A CN201510889210 A CN 201510889210A CN 105306950 A CN105306950 A CN 105306950A
- Authority
- CN
- China
- Prior art keywords
- video
- feedback
- decoder
- encoder
- reconstruction frames
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention discloses a video compression remote transmission system for feeding back coarse quantization reconstructed frames, comprising an information source and an information sink. The information source comprises an information source video encoder, a forward direction transmission transmitter and a reverse direction feedback receiver, wherein the forward direction transmission transmitter and the reverse direction feedback receiver are both in connection with the information source video encoder. The information sink comprises an information sink video decoder, a forward direction transmission receiver, and a reverse direction feedback transmitter, wherein the forward direction transmission receiver and the reverse direction feedback transmitter are both in connection with the information sink video decoder. The forward direction transmission transmitter is in connection with the forward direction transmission receiver; the reverse direction feedback transmitter is in connection with the reverse direction feedback receiver. The video compression remote transmission system feeds back reconstructed video frame information of a remote video decoder on a reverse direction channel, allowing local video decoder reconstructed frames in a local video encoder to be simultaneous with remote video decoder reconstructed frames; the coarse quantization of a transmission data rate is conveniently reduced, and a coarse quantization method is employed for reconstructed frame feedback to reduce a required data rate, thereby reducing the requirement for a decoder operation speed.
Description
Technical field
The present invention relates to the technical field of video compression transmission system, be specifically related to a kind of long-distance video decoder that adopts to the thick method quantizing reconstruction frames of local video condensing encoder feedback, to improve the Subjective and objective qualities of remote decoder video sequence, simultaneously also the feedback of the anti-channel disturbance ability of augmented video compression distance transmission system can slightly quantize the video compression distance transmission system of reconstruction frames.
Background technology
In Bandwidth-Constrained and the channel that is easily disturbed, carry out video communication, need that there is stronger resist miscode ability and high compression performance simultaneously.In order to obtain high compression ratio, the scheme that existing most of Video Codec have employed inter prediction motion compensation effectively reduces time redundancy, then reduce spatial redundancy by spatial alternation, finally entropy code is carried out to result parameter, to reduce the data volume of the rear code stream of compression further.The video standard of this technology is adopted to comprise ITUH.261, MPEG-4, ITUH.263 and ITUH.26L/H.264 standard etc.
In these video standards, video compression encoding algorithm mainly comprises 3 parts: motion prediction, block transform and quantization.A frame of video can adopt two kinds of patterns to carry out compressed encoding, and one is intraframe coding, coding original video frame; Another kind is interframe encode, the motion compensated residual of coding decoded frame.Intraframe coding is mainly used in the initial period of Video coding and works as the moment that video scene there occurs great change.In each mode, be all a block frame being divided into multiple 8*8 size, 4 adjacent brightness blocks and two aberration blocks can synthesize a macro block.After discrete cosine transform (DCT) coefficient to motion-prediction residual quantizes, a local decoder is contained in local video condensing encoder, this local decoder is identical with long-range Video Decoder, it guarantees in ensuing motion-estimation step, coding and decoding video all adopts the video image after reconstruction to carry out motion compensation, thus avoids local video compressed encoding end to adopt original predictive frame and long-distance video decoding end to adopt reconstruction frames to predict the two kinds of asynchronous operation problems brought.If local decoder and long-distance video decoder adopt different reconstruction frames, the motion compensated residual of decoder place decoding can cause a deviation, the phenomenon that will error code be caused further to spread, cause the phenomenon occurring mosaic during video transmission reception in video image, when error code is serious, the smooth and easy broadcasting of receiver, video can be affected.Although the synchronous impact for video transmission quality of reconstruction frames is so important in local decoder and long-distance video decoder, and in the channel of high bit-error, more easily there is the nonsynchronous phenomenon of local and remote reconstruction frames, but current video encoding standard only just takes identical local and remote decoder way to ensure the synchronous of local and remote reconstruction frames, only use the forward channel transmission of video frame information from local video condensing encoder to long-distance video decoder, do not utilize and carry out reconstruction frames feedback from long-distance video decoder to the backward channel in local video condensing encoder direction, the asynchronous problem of the local decoder reconstruction frames therefore in local video condensing encoder and long-distance video decoder reconstructs frame is more serious, directly affect the transmission quality of video compression remote transmission and anti-channel disturbance ability.
In Modern Communication System, both provided the forward channel from encoder transmission machine to Decoder accepts machine direction, the backward channel from Decoder accepts machine to encoder transmission machine was also provided.In the so-called symmetrical service of support, as in the communication system of speech, the capacity of forward and reverse channel can be in a basic balance, and namely backward channel can reach the transmission rate identical with forward channel.Supporting in the communication system as asymmetric traffics such as data services, the capacity ratio of forward channel and backward channel can be regulated according to the different demands of uplink and downlink business, it is generally acknowledged that data service needs larger downstream data rate and less upstream data rate.
Traditional video traffic often needs larger forward channel capacity, namely communication hub node issues video information to the user distributed in systems in which, the feedback control information that on backward channel, general transmission is a small amount of, is generally idle, is not fully used.Particularly when communication network supports multiple multimedia service, the actual data transfer rate difference of backward channel and forward channel is very large, and idle backward channel causes the waste of the resource such as telecommunication system resources and frequency.
Summary of the invention
In order to solve stop asynchronous with long-distance video decoder reconstructs frame because of local decoder reconstruction frames and cause error code diffusion and decoded video quality decline problem, the invention provides a kind of video compression distance transmission system feeding back thick quantification reconstruction frames, the reconstruction video frame information of fed remote Video Decoder on the reverse channel, to make local decoder reconstruction frames in local video condensing encoder and long-distance video decoder reconstructs frame synchronization; Utilize the thick quantification of easy reduction transmitted data rates, reconstruction frames feedback adopts the thick method quantized to reduce the data rate of needs, reduces the requirement to decoder operation speed.
In order to achieve the above object, technical scheme of the present invention is: a kind of video compression distance transmission system feeding back thick quantification reconstruction frames, comprise local information source and the long-range stay of two nights, described information source comprises information source video encoder, forward transmission transmitter and reverse feedback receiver, and forward transmission transmitter is all connected with information source video encoder with reverse feedback receiver; The described stay of two nights comprises stay of two nights Video Decoder, forward transmission receiver and reverse feedback transmitter, and forward transmission receiver is all connected with stay of two nights Video Decoder with reverse feedback transmitter; Described forward transmission transmitter is connected with forward transmission receiver, and reverse feedback transmitter is connected with reverse feedback receiver.
Described information source video compression encoder comprises interframe encoding mode selector in frame, intra encoder, Video Composition encoder, motion compensator, discrete cosine transformer, quantizer, local decoder and feedback reconstruction frames decoder; In described frame, interframe encoding mode selector is connected with video input, intra encoder, motion compensator are all connected with interframe encoding mode selector in frame, motion compensator is connected with discrete cosine transformer, discrete cosine transformer is connected with quantizer, intra encoder, motion compensator and quantizer are all connected with Video Composition encoder, and Video Composition encoder and forward transmit transmitter and be connected; Described local decoder input is connected with feedback reconstruction frames decoder, quantizer respectively, and local decoder output is connected with motion compensator.
Described local decoder comprises information source inverse DCT, information source inverse discrete cosine transformation device and local reconstruction frames, information source inverse DCT is connected with quantizer, information source inverse discrete cosine transformation device is connected with information source inverse DCT, information source inverse discrete cosine transformation device and feedback reconstruction frames decoder are added and export local reconstruction frames buffer to, and local reconstruction frames buffer is connected with motion compensator.
Described stay of two nights Video Decoder comprises video and decomposes thick quantizing encoder in decoder, video sequence follower, intraframe decoder device, stay of two nights inverse DCT, stay of two nights inverse discrete cosine transformation device, inverse motion compensator and feedback frame; Described video decomposes decoder and is connected with forward transmission receiver, video decomposes decoder and is connected with intraframe decoder device, stay of two nights inverse DCT, inverse motion compensator respectively, stay of two nights inverse DCT is connected with stay of two nights inverse discrete cosine transformation device, stay of two nights inverse discrete cosine transformation device is connected with inverse motion compensator, intraframe decoder device is all connected with video sequence follower with inverse motion compensator, video sequence follower is connected with thick quantizing encoder in inverse motion compensator, feedback frame respectively, and in feedback frame, thick quantizing encoder is connected with reverse feedback transmitter.
Described video sequence follower feeds back to thick quantizing encoder in feedback frame the frame of video be combined into, in feedback frame, thick quantizing encoder determines the output data rate of oneself according to the instant capacity of backward channel, in feedback frame, the feedback frame information slightly quantized is sent on backward channel by reverse feedback transmitter by thick quantizing encoder, reverse feedback receiver receives the feedback frame the feedback reconstruction frames decoder be sent in video compression encoder that reverse feedback transmitter sends, the feedback reconstruction frames that feedback reconstruction frames decoder exports and the motion compensated residual that information source inverse discrete cosine transformation device exports and as the local reconstruction frames predicted, be sent to local reconstruction frames buffer, local reconstruction frames is exported to motion compensator by local reconstruction frames buffer.
The invention has the beneficial effects as follows: the backward channel using the information source video encoder from long-range stay of two nights Video Decoder to this locality, the thick reconstruction frames information quantized is fed back to the local decoder in local video compression encoder by long-range stay of two nights Video Decoder, refresh the local reconstruction frames in local decoder, by transmitting the reconstruction frames information of fed remote Video Decoder on the reverse channel, to make local decoder reconstruction frames in local video condensing encoder and long-distance video decoder reconstructs frame synchronization, idle backward channel capacity can be made full use of and improve the quality of transmission of video and the ability of anti-channel disturbance, and do not affect the forward channel capacity of whole communication system, utilize the thick quantification of easy reduction transmitted data rates, reconstruction frames feedback adopts the thick method quantized to reduce the data rate of needs, reduces the requirement to decoder operation speed.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is general principles schematic diagram of the present invention.
Fig. 2 is the concrete structure schematic diagram of information source video encoder in Fig. 1.
Fig. 3 is the concrete structure schematic diagram of stay of two nights Video Decoder in Fig. 1.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not paying the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
Feed back the thick video compression distance transmission system quantizing reconstruction frames, as shown in Figure 1, comprise local information source 100 and the long-range stay of two nights 110.Wherein, information source 100 comprises information source video encoder 101, forward transmission transmitter 102 and reverse feedback receiver 103, and forward transmission transmitter 102 is all connected with information source video encoder 101 with reverse feedback receiver 103.The stay of two nights 110 comprises stay of two nights Video Decoder 111, forward transmission receiver 112 and reverse feedback transmitter 113, and forward transmission receiver 112 is all connected with stay of two nights Video Decoder 111 with reverse feedback transmitter 113.Forward transmission transmitter 102 is connected with forward transmission receiver 112, and reverse feedback transmitter 113 is connected with reverse feedback receiver 103.
Information source video encoder 101 is responsible for video image compression coding, provides compression ratio high as far as possible, to alleviate the demand to forward channel width.The forward transmission transmitter 102 video image compression code stream be responsible for information source video compression encoder 101 exports carries out Channel Coding and Modulation, is then sent to the forward transmission receiver 112 of the long-range stay of two nights 110 by forward transmitting channel.Forward transmission receiver 112 is responsible for receiving the video signal having superposed Noise and Interference pollution through forward channel transmission, and the video signal received is carried out solution mediation channel-decoding, the video after channel-decoding, image information are supplied to stay of two nights Video Decoder 111.Stay of two nights Video Decoder 111 1 aspect is responsible for the compressing video information receiving forward transmission receiver 112 to decompress, recover original video content, owing to can pollute with interference by superimposed noise in forward transmission channel, the video content therefore recovered compares the distortion had to a certain degree with original video content; On the other hand, the decoded video frames recovered is converted into the thick reconstruction frames information quantized and is supplied to reverse feedback transmitter 113 by stay of two nights Video Decoder 111, feeds back with the reverse feedback receiver 103 to information source 100.Reverse feedback transmitter 113 is responsible for reconstruction frames information to be feedback to carry out Channel Coding and Modulation, is then sent to the reverse feedback receiver 103 of information source 100 by reverse feedback channels.Reverse feedback receiver 103 is responsible for carrying out solution mediation channel-decoding to receiving feedback reconstruction frames signal, the feedback reconstruction frames information after channel-decoding is supplied to information source video encoder 101 and uses.The present invention uses the backward channel of the information source video encoder 101 from long-range stay of two nights Video Decoder 111 to this locality, feed back the thick reconstruction frames information quantized in local video compression encoder 111 by long-range stay of two nights Video Decoder 111, refresh local reconstruction frames.
As shown in Figure 2, information source video compression encoder 101 comprises interframe encoding mode selector 202 in frame, intra encoder 203, Video Composition encoder 207, motion compensator 204, discrete cosine transformer 205, quantizer 206, local decoder 210 and feedback reconstruction frames decoder 208.In frame, interframe encoding mode selector 202 is connected with video input 201, intra encoder 203, motion compensator 204 are all connected with interframe encoding mode selector 202 in frame, motion compensator 204 is connected with discrete cosine transformer 205, discrete cosine transformer 205 is connected with quantizer 206, intra encoder 203, motion compensator 204 are all connected with Video Composition encoder 207 with quantizer 206, and Video Composition encoder 207 and forward transmit transmitter 102 and be connected.Local decoder 210 input is connected with feedback reconstruction frames decoder 208, quantizer 206 respectively, and local decoder 210 output is connected with motion compensator 204.
In order to stop asynchronous with long-distance video decoder reconstructs frame because of local decoder reconstruction frames, error code that is that cause spreads and decoded video quality declines, the reconstruction video frame information of the present invention's stay of two nights Video Decoder 111 of fed remote on the reverse channel, with make local decoder 210 reconstruction frames in local information source video encoder 101 and stay of two nights Video Decoder 111 reconstruction frames synchronous.If but vision signal desired rate traditionally feeds back the reconstruction video frame decompressed, very large backward channel capacity will be needed, even far beyond the permission capacity of backward channel, therefore must take measures to reduce the data rate feeding back reconstruction frames, need the data rate determining reasonably to feed back reconstruction frames in other words.Because information source video encoder 101 all needs to use intra-frame encoding mode, this pattern often in the intraframe coding of the starting stage thick quantification that needs transmission one independent of communication, with supplementary reconstruction frames buffering area; When video scene there occurs acute variation, also need to be switched to intra-frame encoding mode and upgrade reconstruction frames buffering area with thick quantization encoding frame.According to the hypothesis of rational up-downgoing channel speed balance, transmission rate during intra-frame encoding mode can think that backward channel can be used for carrying out the data rate of reconstruction frames feedback, also namely backward channel allow capacity, the data rate of this Capacity Ratio original video frame is much lower, therefore reconstruction frames feedback adopts the method for thick quantization encoding to reduce the data rate of needs, thick quantification is a kind of easy reduction transmitted data rates method simultaneously, also reduces the requirement of the arithmetic speed to stay of two nights Video Decoder 111.
Particularly, video input 201 provides unpressed, needs the video source of transmission, and this video source information can be but be not limited to be QCIF form, CIF form, 4CIF form or 16CIF form etc.In frame, interframe encoding mode selector 202 provides the decision-making adopting intraframe coding or inter motion compensation coding, and decision rule can be initial frame or video scene when there is larger changes, employing intraframe coding; Or whether can meet the demands according to the block quality after intraframe coding and inter motion compensation coding.Intra encoder 203 is divided into block by needing the initial frame carrying out intraframe coding, then slightly quantizes the mean value of every block.Particularly, can be 16 grades by the mean quantization of each piece, such quantification will be spacedly distributed between absolute pixel values 52 ~ 228.Block size is determined according to frame target bit rate, that is: frame target bit rate=intra block number × 4.Motion compensator 204 is mainly used for removing inter frame temporal redundancy, particularly, image frame arrangement can be become the block of pixels of 2 × 2 ~ 16 × 16, when locating the particular location of block of pixels, the neighboring area of correspondence position is chosen as region of search in last reconstruction frames, particularly, mean square deviation or mean absolute difference distance metric can be used, also pixel difference class standard can be used, determine the motion vector of corresponding blocks, the predicted value of current block is obtained by suitable motion translation, then deduct from present encoding block and obtain motion compensated residual according to the current block value of reconstructed block prediction.Motion vector is exported to Video Composition encoder 207 by motion compensator 204, motion compensated residual is exported to discrete cosine transformer 205.The motion compensated residual that discrete cosine transformer 205 pairs of motion compensators 204 export carries out discrete cosine transform, exports discrete cosine transform coefficient to quantizer 206.The discrete cosine transform coefficient that quantizer 206 pairs of discrete cosine transformers 205 export carries out optimal quantization.The motion vector that the discrete cosine transform coefficient optimal quantization value of the inter motion compensation residual error that the intraframe coding that intra encoder 203 exports by Video Composition encoder 207, quantizer 206 export and motion compensator 204 export synthesizes the compressed video bitstream of a serial, is sent in forward channel by forward transmission transmitter 102.
In the present invention, in frame, interframe encoding mode selector 202, intra encoder 203, Video Composition encoder 207 form intra-frame encoding mode.In frame, interframe encoding mode selector 202, motion compensator 204, discrete cosine transformer 205, quantizer 206 and Video Composition encoder 207 form interframe encoding mode.In frame, the specific practice of thick quantization encoding is that frame is divided into block by intra encoder 203, then slightly quantizes the mean value of every block.Correspondingly, when feeding back reconstruction frames on the reverse channel, also at long-range stay of two nights Video Decoder 111 place, whole reconstruction frames is divided into block, then the mean value of every block is slightly quantized, again the reconstruction frames after slightly quantizing is fed back in the local decoder 210 of local information source video encoder 101 end by backward channel, upgrade the local reconstruction frames in the local reconstruction frames buffer 2103 in the local decoder 210 of local information source video encoder 101.Local reconstruction frames in local reconstruction frames buffer 2103 in local decoder 210 can regulate as required by the influence degree of feedback reconstruction frames.
Local decoder 210 comprises information source inverse DCT 2101, information source inverse discrete cosine transformation device 2102 and local reconstruction frames buffer 2103, information source inverse DCT 2101 is connected with quantizer 206, information source inverse discrete cosine transformation device 2102 is connected with information source inverse DCT 2101, information source inverse discrete cosine transformation device 2102 and feedback reconstruction frames decoder 208 sum flow to local reconstruction frames buffer 2103, and local reconstruction frames buffer 2103 is connected with motion compensator 204.Feedback reconstruction frames decoder 208 feeds back the anti-thick quantization decoder of the reconstruction frames information come from long-range stay of two nights Video Decoder 111 by reverse feedback receiver 103, reverts to the data format of video frame image.Local decoder 210 is guaranteed in ensuing motion compensation step, and encoding and decoding all adopt the image after reconstruction to carry out motion compensation, thus avoids coding side to adopt primitive frame, and decoding end adopts reconstruction frames to predict the two kinds of asynchronous operation problems brought.
The discrete cosine transform coefficient quantized value inverse of the motion compensated residual that quantizer 206 exports by information source inverse DCT 2101 changes into the discrete cosine transform coefficient value of motion compensated residual.The discrete cosine transform coefficient value of the motion compensated residual that information source inverse DCT 2101 exports by information source inverse discrete cosine transformation device 2102 carries out inverse discrete cosine transformation, exports the motion compensated residual in units of frame.The input of local reconstruction frames buffer 2103 be the feedback reconstruction frames that exports of the motion compensated residual that exports of information source inverse discrete cosine transformation device 2102 and feedback reconstruction frames decoder 208 and, namely it exports is the local reconstruction frames of prediction, and what maintain with long-range stay of two nights Video Decoder 111 reconstruction frames is synchronous.
Motion compensator 204 is the differences calculating the present frame of its input and the predictive frame of local decoder 210 output, i.e. motion compensated residual, and calculating kinematical vector.Motion compensator 204 gives information source discrete cosine transformer 205 motion compensated residual, gives Video Composition encoder 207 after quantification motion vector, and in fact motion compensated residual adds that motion vector is only the difference of notional present frame and previous frame.First frame of video uses intra encoder coding, so complete encoded video streams comprises the coding (being exported by intra encoder 203) of the first frame, the difference (comprising motion vector and motion compensated residual) of follow-up every two interframe, so Video Composition encoder 207 collects three tunnels: the motion compensated residual that the motion vector that intra encoder exports, motion compensator exports of first frame of encoding and discrete cosine transformer export.
Stay of two nights Video Decoder 111 comprises video and decomposes thick quantizing encoder 307 in decoder 301, video sequence follower 302, intraframe decoder device 303, stay of two nights inverse DCT 304, stay of two nights inverse discrete cosine transformation device 305, inverse motion compensator 306 and feedback frame.Video decomposes decoder 301 and is connected with forward transmission receiver 112, video decompose decoder 301 respectively with intraframe decoder device 303, stay of two nights inverse DCT 304, inverse motion compensator 306 is connected, stay of two nights inverse DCT 304 is connected with stay of two nights inverse discrete cosine transformation device 305, stay of two nights inverse discrete cosine transformation device 305 is connected with inverse motion compensator 306, intraframe decoder device 303 is all connected with video sequence follower 302 with inverse motion compensator 306, video sequence follower 302 respectively with inverse motion compensator 306, in feedback frame, thick quantizing encoder 307 is connected, in feedback frame, thick quantizing encoder 307 is connected with reverse feedback transmitter 113.
What the information source video encoder 101 that forward transmission receiver 112 receives from forward channel was sent has superposed interchannel noise and has disturbed the compressed video signal polluted, carry out solution mediation channel-decoding, after removing interchannel noise and interference as far as possible, the compressed video bit stream recovered is sent to video and decompose decoder 301, video decomposes decoder 301 from the video code flow received, decomposites thick quantization encoding information in frame, the quantitative information of the discrete cosine transform coefficient of motion compensation vector information and inter motion compensation residual error, send quantization encoding information thick in frame to intraframe decoder device 303 respectively, send the quantitative information of the discrete cosine transform coefficient of inter motion compensation residual error to stay of two nights inverse DCT 304, motion compensation vector information is sent to inverse motion compensator 306.Intraframe decoder device 303 sends video sequence follower 302 to by after the thick quantitative information decoding in frame; The quantized value of the discrete cosine transform coefficient of inter motion compensation residual error is carried out inverse quantization process by stay of two nights inverse DCT 304, the discrete cosine transform coefficient of the inter motion compensation residual error after inverse quantization is sent to stay of two nights inverse discrete cosine transformation device 305, and between stay of two nights inverse discrete cosine transformation device 305 output frame, motion compensated residual gives inverse motion compensator 306.The video frame image that the decoded video frame image that inverse motion compensator 306 utilizes video sequence follower 302 to export and video decompose the motion compensation vector of decoder 301 output, the inter motion compensation Residual Generation of stay of two nights inverse discrete cosine transformation device 305 output is rebuild, is supplied to video sequence follower 302.The intracoded frame that video sequence follower 302 exports intraframe decoder device 303 and the inter-frame encoding frame that inverse motion compensator 306 exports are combined into complete video flowing.Simultaneously, video sequence follower 302 feeds back to thick quantizing encoder 307 in feedback frame the frame of video be combined into, in feedback frame, thick quantizing encoder 307 determines the output data rate of oneself according to the instant capacity of backward channel, the feedback frame information slightly quantized is sent on backward channel by reverse feedback transmitter 113 by thick quantizing encoder 307 in feedback frame, allows the feedback reconstruction frames decoder 208 in video compression encoder receive.Reverse feedback receiver 103 receive reverse feedback transmitter 113 send feedback frame and the feedback reconstruction frames decoder 208 be sent in video compression encoder 101, the motion compensated residual that the feedback reconstruction frames that feedback reconstruction frames decoder 208 exports exports with information source inverse discrete cosine transformation device 2102 with flow to local reconstruction frames buffer 2103, as the local reconstruction frames of prediction, be sent to motion compensator 204.
Three tunnels that video decomposer 301 decomposes are respectively: the coding of first frame, motion vector and motion compensated residual.The coding of first frame is given intra encoder 303 and is processed, i.e. anti-thick quantification; Motion compensated residual gives the process of this road of stay of two nights inverse DCT 304; Inverse motion compensator 306 is merged into motion vector and motion compensated residual the difference of adjacent two interframe, and add the present frame that video sequence follower 302 exports, the next frame being calculated to be prediction gives video sequence follower 302.Video sequence follower 302 is connected in series first frame and subsequent frame to export and can becomes complete video flowing.
The present invention adopts the thick reconstruction frames information quantized of backward channel feedback of the information source video encoder from long-range stay of two nights Video Decoder to this locality, the synchronization extent of reconstruction frames in local decoder reconstruction frames and long-range stay of two nights Video Decoder in local information source video compression encoder can be improved, thus reach good video long-distance transfer quality and improve the anti-channel disturbance ability of video compression transmission system.Can be used for providing in wireline communication network, cordless communication network and multimedia communication network and carry out real-time or non real-time video compression transmission, also can support that interactive mode communicates with the video compression of two-way full duplex mode, especially can improve the video compression transmission quality in mobile multimedia communication network.
The above; be only the present invention's preferably embodiment, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.
Claims (5)
1. the thick video compression distance transmission system quantizing reconstruction frames of feedback, comprise local information source (100) and the long-range stay of two nights (110), it is characterized in that, described information source (100) comprises information source video encoder (101), forward transmission transmitter (102) and reverse feedback receiver (103), and forward transmission transmitter (102) is all connected with information source video encoder (101) with reverse feedback receiver (103); The described stay of two nights (110) comprises stay of two nights Video Decoder (111), forward transmission receiver (112) and reverse feedback transmitter (113), and forward transmission receiver (112) is all connected with stay of two nights Video Decoder (111) with reverse feedback transmitter (113); Described forward transmission transmitter (102) is connected with forward transmission receiver (112), and reverse feedback transmitter (113) is connected with reverse feedback receiver (103).
2. feedback according to claim 1 slightly quantizes the video compression distance transmission system of reconstruction frames, it is characterized in that, described information source video compression encoder (101) comprises interframe encoding mode selector (202) in frame, intra encoder (203), Video Composition encoder (207), motion compensator (204), discrete cosine transformer (205), quantizer (206), local decoder (210) and feedback reconstruction frames decoder (208), in described frame, interframe encoding mode selector (202) is connected with video input (201), intra encoder (203), motion compensator (204) is all connected with interframe encoding mode selector (202) in frame, motion compensator (204) is connected with discrete cosine transformer (205), discrete cosine transformer (205) is connected with quantizer (206), intra encoder (203), motion compensator (204) is all connected with Video Composition encoder (207) with quantizer (206), Video Composition encoder (207) and forward transmit transmitter (102) and are connected, described local decoder (210) input is connected with feedback reconstruction frames decoder (208), quantizer (206) respectively, and local decoder (210) output is connected with motion compensator (204).
3. feedback according to claim 2 slightly quantizes the video compression distance transmission system of reconstruction frames, it is characterized in that, described local decoder (210) comprises information source inverse DCT (2101), information source inverse discrete cosine transformation device (2102) and local reconstruction frames buffer (2103), information source inverse DCT (2101) is connected with quantizer (206), information source inverse discrete cosine transformation device (2102) is connected with information source inverse DCT (2101), information source inverse discrete cosine transformation device (2102) and feedback reconstruction frames decoder (208) are added and export local reconstruction frames buffer (2103) to, local reconstruction frames buffer (2103) is connected with motion compensator (204).
4. feedback according to claim 1 slightly quantizes the video compression distance transmission system of reconstruction frames, it is characterized in that, described stay of two nights Video Decoder (111) comprises video and decomposes thick quantizing encoder (307) in decoder (301), video sequence follower (302), intraframe decoder device (303), stay of two nights inverse DCT (304), stay of two nights inverse discrete cosine transformation device (305), inverse motion compensator (306) and feedback frame, described video decomposes decoder (301) and is connected with forward transmission receiver (112), video decompose decoder (301) respectively with intraframe decoder device (303), stay of two nights inverse DCT (304), inverse motion compensator (306) is connected, stay of two nights inverse DCT (304) is connected with stay of two nights inverse discrete cosine transformation device (305), stay of two nights inverse discrete cosine transformation device (305) is connected with inverse motion compensator (306), intraframe decoder device (303) is all connected with video sequence follower (302) with inverse motion compensator (306), video sequence follower (302) respectively with inverse motion compensator (306), in feedback frame, thick quantizing encoder (307) is connected, in feedback frame, thick quantizing encoder (307) is connected with reverse feedback transmitter (113).
5. the feedback according to claim 3 or 4 slightly quantizes the video compression distance transmission system of reconstruction frames, it is characterized in that, described video sequence follower (302) feeds back to thick quantizing encoder (307) in feedback frame the frame of video be combined into, in feedback frame, thick quantizing encoder (307) determines the output data rate of oneself according to the instant capacity of backward channel, in feedback frame, the feedback frame information slightly quantized is sent on backward channel by reverse feedback transmitter (113) by thick quantizing encoder (307), reverse feedback receiver (103) receives reverse feedback transmitter (113) feedback frame that sends and the feedback reconstruction frames decoder (208) be sent in video compression encoder (101), the motion compensated residual that feedback reconstruction frames decoder (208) the feedback reconstruction frames that exports and information source inverse discrete cosine transformation device (2102) export and give local reconstruction frames buffer (2103) as what predict, local reconstruction frames buffer (2103) exports local reconstruction frames and is sent to motion compensator (204).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510889210.5A CN105306950B (en) | 2015-12-07 | 2015-12-07 | A kind of video compress distance transmission system for feeding back coarse quantization reconstructed frame |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510889210.5A CN105306950B (en) | 2015-12-07 | 2015-12-07 | A kind of video compress distance transmission system for feeding back coarse quantization reconstructed frame |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105306950A true CN105306950A (en) | 2016-02-03 |
| CN105306950B CN105306950B (en) | 2018-06-15 |
Family
ID=55203625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510889210.5A Expired - Fee Related CN105306950B (en) | 2015-12-07 | 2015-12-07 | A kind of video compress distance transmission system for feeding back coarse quantization reconstructed frame |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105306950B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933703A (en) * | 2016-06-29 | 2016-09-07 | 河南工程学院 | Video transmission system and method capable of feeding back motion compensation residual error information |
| CN112911295A (en) * | 2021-04-16 | 2021-06-04 | 北京杰瑞创通科技有限公司 | Self-adaptive dynamic network packet loss resistant intelligent source coding device and method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101043619A (en) * | 2006-03-24 | 2007-09-26 | 华为技术有限公司 | Error control system and method of video coding |
| CN101127919A (en) * | 2007-09-28 | 2008-02-20 | 中兴通讯股份有限公司 | A video sequence coding method |
| CN101222296A (en) * | 2008-01-31 | 2008-07-16 | 上海交通大学 | Self-adapting transmission method and system in ascending honeycomb video communication |
| CN102160377A (en) * | 2008-09-22 | 2011-08-17 | 史密斯微软件公司 | Video streaming apparatus with quantization and method thereof |
| US20110293009A1 (en) * | 2010-05-27 | 2011-12-01 | Freescale Semiconductor Inc. | Video processing system, computer program product and method for managing a transfer of information between a memory unit and a decoder |
| CN102572440A (en) * | 2012-03-15 | 2012-07-11 | 天津大学 | Multi-viewpoint video transmission method based on depth map and distributed video coding |
| WO2012113763A1 (en) * | 2011-02-24 | 2012-08-30 | Skype | Feedback based reference frame selection for video coding |
| CN205195887U (en) * | 2015-12-07 | 2016-04-27 | 河南工程学院 | Thick video compression teletransmission system that quantifys to rebuild frame of feedback |
-
2015
- 2015-12-07 CN CN201510889210.5A patent/CN105306950B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101043619A (en) * | 2006-03-24 | 2007-09-26 | 华为技术有限公司 | Error control system and method of video coding |
| CN101127919A (en) * | 2007-09-28 | 2008-02-20 | 中兴通讯股份有限公司 | A video sequence coding method |
| CN101222296A (en) * | 2008-01-31 | 2008-07-16 | 上海交通大学 | Self-adapting transmission method and system in ascending honeycomb video communication |
| CN102160377A (en) * | 2008-09-22 | 2011-08-17 | 史密斯微软件公司 | Video streaming apparatus with quantization and method thereof |
| US20110293009A1 (en) * | 2010-05-27 | 2011-12-01 | Freescale Semiconductor Inc. | Video processing system, computer program product and method for managing a transfer of information between a memory unit and a decoder |
| WO2012113763A1 (en) * | 2011-02-24 | 2012-08-30 | Skype | Feedback based reference frame selection for video coding |
| CN102572440A (en) * | 2012-03-15 | 2012-07-11 | 天津大学 | Multi-viewpoint video transmission method based on depth map and distributed video coding |
| CN205195887U (en) * | 2015-12-07 | 2016-04-27 | 河南工程学院 | Thick video compression teletransmission system that quantifys to rebuild frame of feedback |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105933703A (en) * | 2016-06-29 | 2016-09-07 | 河南工程学院 | Video transmission system and method capable of feeding back motion compensation residual error information |
| CN112911295A (en) * | 2021-04-16 | 2021-06-04 | 北京杰瑞创通科技有限公司 | Self-adaptive dynamic network packet loss resistant intelligent source coding device and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105306950B (en) | 2018-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100471269C (en) | Spatial scalable compression | |
| Assuncao et al. | Post-processing of MPEG2 coded video for transmission at lower bit rates | |
| JP5118127B2 (en) | Adaptive encoder assisted frame rate upconversion | |
| CN1856111B (en) | Video signal coding/decoding method, coder/decoder and related devices | |
| CN102318202B (en) | System and method for transcoding between scalable and non-scalable video codecs | |
| CN101001371B (en) | Method of video transcoding and its device | |
| CN103957389A (en) | 3G video transmission method and system based on compression sensing | |
| CN102724498A (en) | Methods and device for data alignment with time domain boundary | |
| CN101622880A (en) | Effective video block mode in the second leg video coding changes | |
| CN101421936A (en) | System and method for providing error resilience, random access and rate control in scalable video communications | |
| CN103108180B (en) | A kind of method based on infrastructure ability and conditions present determination Video coding sub-block size and device thereof | |
| CN105103563A (en) | Method and apparatus for quantization matrix signaling and representation in scalable video coding | |
| CN101888550A (en) | Encoding method and device of quantization parameters in slice head information | |
| CN1938728A (en) | Method and apparatus for encoding a picture sequence using predicted and non-predicted pictures which each include multiple macroblocks | |
| CN101888547B (en) | Direct mode selection based H.264/AVC (Advanced Video Coding) fast transcoding method and device | |
| CN100502503C (en) | Code-transferring system and method for simultaneously realizing multiple code flow output | |
| CN105306950A (en) | Video compression remote transmission system for feeding back coarse quantization reconstructed frames | |
| CN102082919A (en) | Digital video matrix | |
| CN205195887U (en) | Thick video compression teletransmission system that quantifys to rebuild frame of feedback | |
| JP2005535219A (en) | Method and apparatus for performing multiple description motion compensation using hybrid prediction code | |
| Sun et al. | The improved SP frame coding technique for the JVT standard | |
| US20060093031A1 (en) | Method and apparatus for performing multiple description motion compensation using hybrid predictive codes | |
| CN105933703A (en) | Video transmission system and method capable of feeding back motion compensation residual error information | |
| CN103051926A (en) | Streaming transcoder with adaptive upstream and downstream transcode coordination | |
| CN205883463U (en) | Video transmission system of feedback motion compensation residual error information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180615 Termination date: 20201207 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |