CN103379351A - Video processing method and video processing device - Google Patents

Video processing method and video processing device Download PDF

Info

Publication number
CN103379351A
CN103379351A CN2012101326822A CN201210132682A CN103379351A CN 103379351 A CN103379351 A CN 103379351A CN 2012101326822 A CN2012101326822 A CN 2012101326822A CN 201210132682 A CN201210132682 A CN 201210132682A CN 103379351 A CN103379351 A CN 103379351A
Authority
CN
China
Prior art keywords
pixel
image
video data
resolution
synchronization frame
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
Application number
CN2012101326822A
Other languages
Chinese (zh)
Other versions
CN103379351B (en
Inventor
张博洋
刘琚
葛菁
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Mobile Group Shandong Co Ltd
Original Assignee
China Mobile Group Shandong Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by China Mobile Group Shandong Co Ltd filed Critical China Mobile Group Shandong Co Ltd
Priority to CN201210132682.2A priority Critical patent/CN103379351B/en
Publication of CN103379351A publication Critical patent/CN103379351A/en
Application granted granted Critical
Publication of CN103379351B publication Critical patent/CN103379351B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

The invention discloses a video processing method and a video processing device. The video processing method and the video processing device are used for solving the problem that the data volume of the three-dimension video in the prior art is excessive. The method receives first video data with high frame rate and low resolution and second video data with low frame rate and high resolution. The resolution of the first video data is recovered according to the second video data with the high resolution to obtain the first video data with the high frame rate and the high resolution. The frame rate of the second video data is recovered according to the first video data with the high frame rate and the high resolution to obtain the second video data with the high frame rate and the high resolution. Sequential processing is performed based on the obtained first video data with the high frame rate and the high resolution and the obtained second video data with the high frame rate and the high resolution. With the method, the resolution of one path of the video data and the frame rate of the other path of the video data are reduced when the video is collected, and in addition, the two paths of the video data can be recovered to be with high frame and high resolution. Thus, displaying quality of the video is ensured and meanwhile, the data volume of the video is reduced.

Description

A kind of method for processing video frequency and device
Technical field
The present invention relates to communication technical field, relate in particular to a kind of method for processing video frequency and device.
Background technology
In recent years, the application of three-dimensional (Three Dimensions, 3D) video has obtained tremendous development, and the application scenarios that need to carry out the collection of 3D video also gets more and more.
In the prior art, in order to improve the display quality of 3D video as far as possible, need to gather the higher and identical frame per second of two-way and resolution video data when gathering the 3D video.For example, for an object, gather the 3D video of this object, just need respectively two of this object different angles, gather the video data of this object with higher and identical frame per second and resolution, again the two-path video data that obtain are carried out follow-up processing, and then obtain the 3D video of this object.
Therefore, the data volume of 3D video will much larger than the data volume of single view video, also just need to take more storage resources in the prior art when storage 3D video.
Summary of the invention
The embodiment of the invention provides a kind of method for processing video frequency and device, and is excessive in order to the data volume that solves 3D video in the prior art, need to take the problem of a large amount of storage resources.
A kind of method for processing video frequency that the embodiment of the invention provides comprises:
Receive the first video data and the second video data, the frame per second of described the first video data is higher than the frame per second of described the second video data, and the resolution of the image in described the first video data is lower than the resolution of the image in described the second video data;
Synchronization frame image corresponding to image definite and in described the second video data in described the first video data, based on the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence;
According to the synchronization frame image after processing, be identical with the resolution of synchronization frame image after the processing with the resolution processes of the async framing image in described the first video data, the first video data after obtaining processing;
According to the first video data after the described processing, determine in described the second video data the disappearance two field picture with respect to the first video data after the described processing, the disappearance two field picture of determining is inserted in described the second video data the second video data after obtaining processing;
Carry out subsequent treatment based on the first video data after the described processing and the second video data after the described processing.
A kind of video process apparatus that the embodiment of the invention provides comprises:
Receiver module, be used for receiving the first video data and the second video data, the frame per second of described the first video data is higher than the frame per second of described the second video data, and the resolution of the image in described the first video data is lower than the resolution of the image in described the second video data;
Resolution is recovered module, be used for synchronization frame image corresponding to image definite at described the first video data and in described the second video data, based on the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence, according to the synchronization frame image after processing, with the resolution processes of the async framing image in described the first video data be with process after the resolution of synchronization frame image identical, the first video data after obtaining processing;
Frame per second is recovered module, be used for according to the first video data after the described processing, determine in described the second video data the disappearance two field picture with respect to the first video data after the described processing, the disappearance two field picture of determining is inserted in described the second video data the second video data after obtaining processing;
Processing module is used for carrying out subsequent treatment based on the first video data after the described processing and the second video data after the described processing.
The embodiment of the invention provides a kind of method for processing video frequency and device, the method receives the first video data of high frame per second low resolution, and high-resolution the second video data of low frame per second, and according to the every vertical frame dimension image in different resolution in the second video data, recover the resolution of each image in the first video data, obtain high-resolution the first video data of high frame per second, again according to high-resolution the first video data of high frame per second that obtains, recover the frame per second of the second video data, obtain high-resolution the second video data of high frame per second, carry out subsequent treatment based on high-resolution the first video data of the high frame per second that obtains and high-resolution the second video data of high frame per second.Pass through said method, can be when gathering the 3D video, reduce the resolution of one road video data and the frame per second of another road video data, and when the two-path video data that processing collects, the two-path video data are reverted to high frame per second and high-resolution, therefore can in the display quality that guarantees the 3D video, reduce the data volume of 3D video, save storage resources.
Description of drawings
The video processing procedure that Fig. 1 provides for the embodiment of the invention;
Collection the first video data and the second video data schematic diagram that Fig. 2 provides for the embodiment of the invention;
High-resolution the first video data of high frame per second and high-resolution the second video data contrast of low frame per second schematic diagram that Fig. 3 provides for the embodiment of the invention;
High-resolution the first video data of high frame per second and high-resolution the second video data contrast of high frame per second schematic diagram that Fig. 4 provides for the embodiment of the invention;
The detailed process that the first video data of high frame per second low resolution is reverted to high-resolution the first video data of high frame per second that Fig. 5 provides for the embodiment of the invention;
Fig. 6 will hang down the detailed process that high-resolution the second video data of frame per second reverts to high-resolution the second video data of high frame per second for what the embodiment of the invention provided;
The schematic diagram of the disappearance vertical frame dimension image in different resolution between determining of providing of the embodiment of the invention high-definition picture that two frames are adjacent in the second video data is provided Fig. 7;
The video process apparatus structural representation that Fig. 8 provides for the embodiment of the invention.
Embodiment
As everyone knows, frame per second and resolution are two key factors that affect video display quality.Frame per second refers to that per second shows the frame number of image, and frame per second is higher, and the fluency of video and fidelity are also just higher.Resolution refers to the pixel of a two field picture, and resolution is higher, and the details that image comprises is also just more, and its fineness is also just higher.But for one road video data, frame per second and resolution affect the data volume size of this video data equally, and frame per second is higher, and its data volume is larger, and resolution is higher, and its data volume is larger.In order to improve the display quality of 3D video, when gathering the 3D video, need to gather the video data of the higher and identical frame per second of two-way and resolution in the prior art, can make undoubtedly the data volume of 3D video excessive.
In embodiments of the present invention, in order to reduce the data volume of 3D video, when gathering the 3D video, the two-path video data of collection are: high-resolution the second video data of the first video data of high frame per second low resolution and low frame per second.When the two-path video data that processing collects, recover the resolution of the first video data by the second video data, again according to high-resolution the first video data of high frame per second that recovers, recover the frame per second of the second video data, obtain the high-resolution video data of the high frame per second of two-way, to reach in the display quality that guarantees the 3D video, reduce the purpose of the data volume of 3D video.
Below in conjunction with Figure of description, the embodiment of the invention is described in detail.
Fig. 1 is the video processing procedure that the embodiment of the invention provides, and specifically may further comprise the steps:
S101: receive the first video data and the second video data.
In embodiments of the present invention, the first video data that receives and the second video data are for same collection target, the two-path video data that collect in the different angles of this collection target, and, the frame per second of the first video data is higher than the frame per second of the second video data, and the resolution of the image in the first video data is lower than the resolution of the image in the second video data.Also be, when gathering the 3D video, adopt high frame per second low resolution to gather the first video data, adopt low frame per second high-resolution acquisition the second video data, the two-path video data comparatively speaking, the first video data is the first video data of high frame per second low resolution, and the second video data is low high-resolution the second video data of frame per second, as shown in Figure 2.
Collection the first video data and the second video data schematic diagram that Fig. 2 provides for the embodiment of the invention, in Fig. 2, for gathering target, gather the first video data by collecting device A at the LOOK LEFT that gathers target, gather the second video data by collecting device B at the LOOK RIGHT that gathers target, wherein, collecting device A and B can be the equipment such as video camera.In Fig. 2, because the first video data that collecting device A gathers is the video data of high frame per second low resolution, the second video data that collecting device B gathers is the low high-resolution video data of frame per second, therefore within the identical time, the frame number of the image that comprises in the first video data is more than the frame number of the image that comprises in the second video data, and the size of every two field picture is greater than the size of every two field picture in the first video data in the second video data.
After collecting the first video data and the second video data, then these two-path video data are sent to video process apparatus and process.
S102: synchronization frame image corresponding to image definite and in the second video data in the first video data, based on the image in the second video data, be identical with the resolution of this image in the second video data with the resolution processes of the synchronization frame image of correspondence.
Image in first video data that below will gather is called low-resolution image, and the image in the second video data that gathers is called high-definition picture.Also be, for the every vertical frame dimension image in different resolution in this second video data, in this first video data, determine the synchronization frame low-resolution image corresponding with this high-definition picture, based on this high-definition picture this synchronization frame low-resolution image is treated to the synchronization frame high-definition picture.
In embodiments of the present invention, because the frame per second of the first video data is higher than the frame per second of the second video data, if therefore the two-path video data are shown separately, then within the identical time, the frame number of the image that the first video data shows will be more than the frame number of the second video data demonstration.And when these two-path video data are shown simultaneously, also namely begin to show the 1st two field picture in the two-path video data at synchronization, when 2nd frame of projection in the second video data, the image that the first video data is being shown is synchronization frame image corresponding to the 2nd true image in the second video data.Also namely, for the every vertical frame dimension image in different resolution in the second video data, determine projection to this high-resolution duration, definite image of showing when the first video data is shown this duration is as synchronization frame low-resolution image corresponding to this high-definition picture.
Continuation is take Fig. 2 as example, and in Fig. 2, the synchronization frame low-resolution image corresponding with every vertical frame dimension image in different resolution in the second video data is the image of shade sign in the first video data.
Concrete, the every vertical frame dimension image in different resolution in this second video data can adopt formula
Figure BDA0000158998870000051
Determine synchronization frame low-resolution image corresponding with this high-definition picture in the first video data, wherein a is the frame per second of the first video data, it also is described high frame per second, b is the frame per second of the second video data, it also is described low frame per second, M represents the M two field picture in the second video data, and N represents that the N two field picture in the first video data is synchronization frame low-resolution image corresponding to M two field picture in the second video data
Figure BDA0000158998870000061
Expression is in the mode that rounds up pair
Figure BDA0000158998870000062
Round.
After having determined synchronization frame low-resolution image corresponding to high-definition picture in the second video data, can be identical with the resolution of high-definition picture in the second video data with the resolution processes of the synchronization frame low-resolution image of correspondence based on this high-definition picture, also namely recover the resolution of synchronization frame low-resolution image in the first video data, the synchronization frame low-resolution image is treated to the synchronization frame high-definition picture.
S103: according to the synchronization frame image after processing, be identical with the resolution of synchronization frame image after the processing with the resolution processes of the async framing image in the first video data, the first video data after obtaining processing.
In embodiments of the present invention, after each synchronization frame low-resolution image in the first video data reverted to the synchronization frame high-definition picture, then can be further according to each the synchronization frame high-definition picture that recovers, each async framing low-resolution image (Non-alignment Frame low-resolution image) in the first video data is reverted to the async framing high-definition picture, and then the first video data after obtaining processing, the first video data after the processing is high-resolution the first video data of high frame per second.
Continuation is according to the every vertical frame dimension image in different resolution in the second video data among the step S102 take Fig. 2 as example, and every frame in the first video data is reverted to the synchronization frame high-definition picture with the synchronization frame low-resolution image that shade identifies.Then be according to the synchronization frame high-definition picture that has recovered among the step S103, the every frame async framing low-resolution image that does not identify with shade in the first video data is reverted to the async framing high-definition picture.At this moment, the every frame low-resolution image in the first video data is all reverted to high-definition picture, therefore obtained high-resolution the first video data of high frame per second.
High-resolution the first video data of the high frame per second that obtains as shown in Figure 3.High-resolution the first video data of high frame per second and high-resolution the second video data contrast of low frame per second schematic diagram that Fig. 3 provides for the embodiment of the invention, as shown in Figure 3, because the first video data with high frame per second low resolution has reverted to high-resolution the first video data of high frame per second, therefore every two field picture measure-alike in the size of the every two field picture in high-resolution the first video data of high frame per second and the second video data all is the size of this high-definition picture.In Fig. 3, synchronization frame high-definition picture corresponding with every vertical frame dimension image in different resolution in the second video data in high-resolution the first video data of the high frame per second that obtains is still with shadow representation.
S104: according to the first video data after processing, determine in the second video data with respect to the disappearance two field picture of the first video data after processing the disappearance two field picture of determining to be inserted in the second video data the second video data after obtaining processing.
Because the frame per second of the second video data is lower than the frame per second of high-resolution the first video data of high frame per second that obtains, so the frame number of the image that comprises in high-resolution the first video data of high frame per second in the same time will be more than the frame number of the image that comprises in the second video data.As shown in Figure 3, when showing next synchronization frame high-definition picture from a synchronization frame high-definition picture of high-resolution the first video data of high frame per second, comprise this two synchronization frame high-definition pictures, show altogether 5 two field pictures, and in the identical time, the second video data has only been shown corresponding two vertical frame dimension image in different resolution, therefore, lack 3 two field pictures with respect to the first video data between this two vertical frame dimensions image in different resolution in the second video data, be disappearance vertical frame dimension image in different resolution.
In embodiments of the present invention, according to high-resolution the first video data of the high frame per second that obtains, determine in this second video data the disappearance vertical frame dimension image in different resolution with respect to high-resolution the first video data of high frame per second, the disappearance vertical frame dimension image in different resolution of determining is inserted in this second video data, thereby can recover the frame per second of the second video data, to hang down high-resolution the second video data of frame per second and revert to high-resolution the second video data of high frame per second, the second video data after also namely processing is high-resolution the second video data of high frame per second, as shown in Figure 4.
High-resolution the first video data of high frame per second and high-resolution the second video data contrast of high frame per second schematic diagram that Fig. 4 provides for the embodiment of the invention, in Fig. 4, because the first video data and the second video data have all reverted to high frame per second high-resolution, the frame number of the image that therefore the two-path video data comprise in the same time is identical, and the resolution of the image that the two-path video data comprise is also identical.
S105: carry out subsequent treatment based on high-resolution the first video data of the high frame per second that obtains and high-resolution the second video data of high frame per second.
In said process, video process apparatus receives the first video data of high frame per second low resolution, and high-resolution the second video data of low frame per second, and according to the every vertical frame dimension image in different resolution in the second video data, recover the resolution of each image in the first video data, obtain high-resolution the first video data of high frame per second, again according to high-resolution the first video data of high frame per second that obtains, recover the frame per second of the second video data, obtain high-resolution the second video data of high frame per second, carry out subsequent treatment based on high-resolution the first video data of the high frame per second that obtains and high-resolution the second video data of high frame per second.Pass through said method, can be when gathering the 3D video, reduce the resolution of one road video data and the frame per second of another road video data, and when the two-path video data that processing collects, the two-path video data are reverted to high frame per second high-resolution, therefore can in the display quality that guarantees the 3D video, reduce the data volume of 3D video, save storage resources.And because the embodiment of the invention has reduced the data volume of 3D video, therefore when this 3D video of transmission, do not need very large bandwidth yet, therefore saved network bandwidth resources yet.
In process shown in Figure 2, step S102 and S103 are the processes of resolution of recovering the first video data of high frame per second low resolution, and the first video data that also is about to high frame per second low resolution reverts to the process of high-resolution the first video data of high frame per second.Step S104 is the process of recovering the frame per second of low high-resolution the second video data of frame per second, also is about to the process that low high-resolution the second video data of frame per second reverts to high-resolution the second video data of high frame per second.The below is elaborated to these two processes respectively.
When the resolution of the first video data that recovers high frame per second low resolution, need first the low respectively rate image of the synchronization frame in the first video data to be reverted to the synchronization frame high-definition picture, again the Non-alignment Frame low-resolution image is reverted to the Non-alignment Frame high-definition picture.
In the embodiment of the present application, the method that synchronization frame low-resolution image in the first video data is reverted to the synchronization frame high-definition picture is specially: the synchronization frame image (synchronization frame low-resolution image) corresponding with the image (high-definition picture in the second video data) in the second video data carried out up-sampling process, the image in the size of the synchronization frame image behind the up-sampling and this second video data measure-alike; Determine image in this second video data and the parallax of the synchronization frame image behind the up-sampling, based on the parallax of determining and the image in this second video data, be identical with the resolution of image in this second video data with the resolution processes of the synchronization frame image behind the up-sampling.
In the above-mentioned process that the synchronization frame low-resolution image is reverted to the synchronization frame high-definition picture, because the first video data and the second video data are to be collected in the different angles of same collection target respectively by two collecting devices, therefore for the high-definition picture in the second video data, this gathers the relative position of target in this high-definition picture, with this gather target in the first video data with synchronization frame low-resolution image corresponding to this high-definition picture in relative position different.Describe as an example of Fig. 2 example, the first video data is that collecting device A gathers at the LOOK LEFT that gathers target, and the second video data is that collecting device B is in the LOOK RIGHT collection of collection target, therefore should gather the relative position in a certain two field picture of target in the second video data, to gather target relative position in the corresponding synchronization frame image in the first video data different with this.The difference that gathers target relative position in this two two field picture is exactly the parallax of this two two field picture.
And in order to determine the parallax of this high-definition picture in this synchronization frame low-resolution image and the second video data in the first video data, then needing first the size of synchronization frame low-resolution image to be carried out up-sampling processes, make the measure-alike of the size of the synchronization frame low-resolution image behind the up-sampling and high-definition picture, wherein, the synchronization frame low-resolution image is carried out up-sampling and process specifically interpolation up-sampling processing.The size with the synchronization frame low-resolution image be treated to high-definition picture measure-alike after, just can determine the parallax of the synchronization frame low-resolution image behind this high-definition picture and the up-sampling.
In the embodiment of the invention in order to improve the accuracy of definite parallax, when this image in determining the second video data and the parallax of the synchronization frame image behind the up-sampling, first the image (high-definition picture) in this second video data being carried out down-sampling processes, corresponding synchronization frame image (synchronization frame low-resolution image) is measure-alike in the first video data of the size of the image behind the down-sampling and collection, again the image behind this down-sampling is carried out up-sampling and process, pass through successively the measure-alike of the size of the image behind down-sampling and the up-sampling and the synchronization frame image behind the up-sampling; Determine that by the method for image calibration and Stereo matching this passes through image behind down-sampling and the up-sampling and the parallax of the synchronization frame image behind this up-sampling successively, as the image in this second video data of determining and the parallax of the synchronization frame image behind the up-sampling.Wherein, the purpose of first this high-definition picture in the second video data being carried out the down-sampling processing is in order to reduce the resolution of this high-definition picture, make its resolution with the synchronization frame low-resolution image identical, the image behind the down-sampling being carried out that up-sampling processes is the size that again reverts to the original high resolution image for the size of the image that will reduce resolution again, also is that size is with identical through the synchronization frame low-resolution image after the up-sampling processing.Like this, through just having had identical size and identical resolution through the image behind down-sampling and the up-sampling successively in the synchronization frame low-resolution image behind the up-sampling and the second video data, determine that in the situation of same size and equal resolution the parallax of two two field pictures can improve the accuracy of definite parallax in the first video data.
Adopt said method to determine the synchronization frame low-resolution image behind the up-sampling and pass through successively down-sampling and up-sampling after the parallax of high-definition picture after, based on original this high-definition picture (not passing through the high-definition picture of down-sampling and up-sampling) in the parallax of determining and the second video data, the resolution processes of the synchronization frame image behind the up-sampling is specially for the method identical with the resolution of original this high-definition picture in the second video data: the pixel value of determining to pass through successively each pixel in the image behind down-sampling and the up-sampling, the difference of the pixel value of corresponding each pixel in the image after determining the pixel value of each pixel in the image (not passing through the high-definition picture of down-sampling and up-sampling) in the second video data and passing through successively down-sampling and up-sampling is as the high fdrequency component of each pixel in the image in this second video data; Based on the parallax of determining, determine in the image in this second video data the corresponding relation of each pixel in the synchronization frame image (the synchronization frame low-resolution image behind the up-sampling) behind each pixel and this up-sampling; According to the corresponding relation of determining, the pixel value of each pixel in the synchronization frame image behind this up-sampling is increased the high fdrequency component of pixel corresponding in the image in this second video data, obtain the identical synchronization frame image of resolution of image original in resolution and this second video, namely obtain the synchronization frame high-definition picture.
In said process, be successively the low frequency component of each pixel in the high-definition picture original in the second video data through the pixel value of each pixel in the image behind down-sampling and the up-sampling, and in this high-definition picture the pixel value of each pixel be high fdrequency component and low frequency component and value, therefore, the pixel value of each pixel and the high fdrequency component that is successively each pixel in this high-definition picture through the difference of the pixel value of corresponding each pixel in the image behind down-sampling and the up-sampling in this high-definition picture.Because the synchronization frame low-resolution image behind the up-sampling and the parallax of this high-definition picture are determined, therefore can determine accordingly in this high-definition picture the corresponding relation of each pixel in the synchronization frame low-resolution image behind each pixel and up-sampling, again based on this corresponding relation, the high fdrequency component of each pixel in this high-definition picture is added in the synchronization frame low-resolution image behind the up-sampling on the corresponding pixel, synchronization frame image after can obtaining processing, the synchronization frame image after the processing is the synchronization frame high-definition picture.Certainly, also can determine by the method for low-pass filtering and high-pass filtering low frequency component and the high fdrequency component of each pixel in the high-definition picture original in the second video data.
So far, the low respectively rate image of each synchronization frame in the first video data is reverted to the synchronization frame high-definition picture, the following describes the process of the resolution of recovering the async framing low-resolution image in the first video data.
In recovering the first video data during the resolution of each async framing low-resolution image, be based on that the synchronization frame high-definition picture that recovered in the first video data recovers in the embodiment of the invention.Be specially: async framing image (async framing low-resolution image) is carried out up-sampling processes, the size of the async framing image behind the up-sampling with process after synchronization frame image (synchronization frame high-definition picture) measure-alike; Async framing image behind the up-sampling is divided into several pixel macroblock, carry out for each pixel macroblock of dividing: by the method for estimation, determine and the immediate pixel macroblock of this pixel macroblock in the synchronization frame image after the processing of appointment, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining.Also namely, when recovering the resolution of async framing low-resolution image, according to the synchronization frame high-definition picture of appointment, recover respectively the resolution of each pixel macroblock in this async framing low-resolution image, obtain the async framing high-definition picture.
Wherein, the synchronization frame image after the processing of appointment is: be positioned at the synchronization frame image after the processing nearest before this async framing image in the first video data, and be arranged in synchronization frame image after the processing nearest after this async framing image at least one.With the immediate pixel macroblock of pixel macroblock of dividing be: the pixel macroblock of the least absolute error of the pixel value of each pixel and value or least mean-square error value minimum in the pixel value of each pixel that comprises and the pixel macroblock of division.
Also be, after each synchronization frame low-resolution image reverts to the synchronization frame high-definition picture in the first video data, for the async framing low-resolution image in the first video data, first the size up-sampling of this async framing low-resolution image is treated to measure-alike with the synchronization frame high-definition picture, again the async framing low-resolution image behind the up-sampling is divided into several pixel macroblock, for each pixel macroblock of dividing, method by estimation, with the nearest synchronization frame high-definition picture in these async framing low-resolution image both sides in, determine and the immediate pixel macroblock of this pixel macroblock, the high fdrequency component of each pixel in the immediate pixel macroblock of determining is added in this pixel macroblock of dividing on corresponding each pixel.After each pixel macroblock of dividing carried out aforesaid operations, this async framing low-resolution image can be reverted to the async framing high-definition picture.
In embodiments of the present invention, in order to improve the accuracy of the resolution of recovering the async framing low-resolution image, to improve the display quality of 3D video, in said process, for each pixel macroblock of dividing, the pixel value of each pixel in the pixel macroblock of dividing is increased before the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining, also will be according to the pixel macroblock of dividing and definite pixel macroblock immediate with it, determine that the pixel macroblock of division is with respect to the motion vector of pixel macroblock immediate with it, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of determining to divide according to the motion vector of determining, pixel value according to each pixel in the pixel macroblock of dividing, the pixel value of each pixel in this immediate pixel macroblock, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of dividing is determined the pixel macroblock of division and the variance rate of this immediate pixel macroblock.
When determining variance rate, can pass through formula
Figure BDA0000158998870000121
Determine this pixel macroblock of division and the variance rate of this immediate pixel macroblock, wherein, T (i, j) be the pixel value of the pixel of the capable j row of i in this pixel macroblock of dividing, R (i, j) is the pixel value of the pixel of the capable j row of i in this immediate pixel macroblock, wherein, the pixel that the capable j of i is listed as in the pixel of the capable j row of i and this immediate pixel macroblock in this pixel macroblock of dividing has corresponding relation, SAD Norm(T, R) is this pixel macroblock of definite division and the variance rate of this immediate pixel macroblock.As seen, the variance rate of determining is less, and this pixel macroblock of illustrated divisions is more similar to this immediate pixel macroblock.
After having determined variance rate, the pixel value of each pixel increases the high fdrequency component of each pixel in definite pixel macroblock immediate with it in this pixel macroblock that then can will divide according to the variance rate of determining.Specifically can be divided into following two kinds of situations:
Situation one: when the variance rate of determining when setting the variance rate threshold value, this pixel macroblock of illustrated divisions is very high with this immediate pixel macroblock similarity, therefore can directly the pixel value of each pixel in this pixel macroblock of dividing be increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining;
Situation two: when the variance rate of determining is not less than setting variance rate threshold value, this pixel macroblock of illustrated divisions is not high with this immediate pixel macroblock similarity, therefore need according to this immediate pixel macroblock, and other pixel macroblock adjacent with this immediate pixel macroblock are recovered the resolution of this pixel macroblock of division jointly in the synchronization frame image after the processing at this immediate pixel macroblock place (synchronization frame high-definition picture).Be specially: in the synchronization frame high-definition picture at this immediate pixel macroblock place, determine each neighbor macro block adjacent with this immediate pixel macroblock; For each pixel in this immediate pixel macroblock, according to the position of this pixel in this immediate pixel macroblock, determine the high fdrequency component of each pixel on the relevant position in each neighbor macro block, according to the weights of setting for this pixel in this pixel macroblock that connects recently, and be the weights that each pixel on the relevant position is set in each neighbor macro block, weighted average with the high fdrequency component of each pixel on the relevant position in the high fdrequency component of this pixel and each the neighbor macro block redefines the high fdrequency component into this pixel in this immediate pixel macroblock; The pixel value of each pixel in this pixel macroblock of dividing is increased the high fdrequency component of each pixel that redefines in the pixel macroblock immediate with it.
Wherein, in the synchronization frame high-definition picture at this immediate pixel macroblock place, each neighbor macro block adjacent with this immediate pixel macroblock of determining comprises: the pixel macroblock that is positioned at this immediate pixel macroblock upside, downside, left side, right side in the synchronization frame high-definition picture at this immediate pixel macroblock place.Each pixel in this immediate pixel macroblock can adopt formula
Figure BDA0000158998870000131
The high fdrequency component of this pixel that redefines, wherein, n represents n pixel macroblock of this immediate pixel macroblock and these 5 pixel macroblock of each neighbor macro block (being arranged in 4 pixel macroblock on this immediate pixel macroblock upside, downside, left side, right side), W n(i, j) is expressed as the weights of the pixel setting of the capable j row of i in n the pixel macroblock, B nThe high fdrequency component of the pixel of the capable j row of i in n pixel macroblock of (i, j) expression, B New(i, j) is the high fdrequency component of the pixel of the capable j row of i in this immediate pixel macroblock that redefines.
After adopting said method to redefine the pixel value of each pixel in this immediate pixel macroblock, then the pixel value of each pixel in the pixel macroblock of dividing can be increased the high fdrequency component of each pixel that redefines in this immediate pixel macroblock, with the resolution of this pixel macroblock of recovering to divide.
Certainly, after having determined variance rate, when the pixel value of each pixel increases the high fdrequency component of each pixel in definite pixel macroblock immediate with it in this pixel macroblock that will divide according to the variance rate of determining, also can set two variance rate threshold values, i.e. the first variance rate threshold value and the second variance rate threshold value, wherein, the first variance rate threshold value is less than the second variance rate threshold value.When the variance rate of determining during less than the first variance rate threshold value, carry out corresponding operating according to above-mentioned situation one.When the variance rate of determining is not less than the first variance rate threshold value and during less than the second variance rate threshold value, carries out corresponding operatings according to above-mentioned situation two.When the variance rate of determining is not less than the second variance rate threshold value, this pixel macroblock of illustrated divisions is lower with the similarity of this immediate pixel macroblock of determining, even jointly recover the resolution of this pixel macroblock of dividing by this immediate pixel macroblock and each neighbor macro block adjacent with this immediate pixel macroblock, its recovery effects is not good yet, and the pixel macroblock of this pixel macroblock that therefore directly will divide after as recovery resolution gets final product.
Fig. 5 is the detailed process that the first video data of high frame per second low resolution is reverted to high-resolution the first video data of high frame per second that the embodiment of the invention provides, and specifically may further comprise the steps:
S501: for the every vertical frame dimension image in different resolution in this second video data, in the first video data of this high frame per second low resolution, determine the synchronization frame low-resolution image corresponding with this high-definition picture.
S502: this synchronization frame low-resolution image is carried out up-sampling process, the size of the synchronization frame low-resolution image behind the up-sampling and this high-definition picture measure-alike.
S503: this high-definition picture is carried out down-sampling process, the size of the high-definition picture behind the down-sampling is measure-alike with the synchronization frame low-resolution image that passes through the up-sampling processing.
S504: the high-definition picture behind this down-sampling is carried out up-sampling process, pass through successively the measure-alike of the size of the high-definition picture behind down-sampling and the up-sampling and the synchronization frame low-resolution image behind the up-sampling.
S505: the method by image calibration and Stereo matching determines to pass through successively the high-definition picture and the parallax of the synchronization frame low-resolution image behind the up-sampling behind down-sampling and the up-sampling, as this definite high-definition picture and the parallax of the synchronization frame low-resolution image behind the up-sampling.
S506: the pixel value of determining to pass through successively each pixel in the high-definition picture behind down-sampling and the up-sampling, the difference of the pixel value of corresponding each pixel in the high-definition picture after determining the pixel value of each pixel in undressed this high-definition picture and passing through successively down-sampling and up-sampling is as the high fdrequency component of each pixel in undressed this high-definition picture.
S507: based on the parallax of determining, determine in undressed this high-definition picture the corresponding relation of each pixel in the synchronization frame low-resolution image behind each pixel and up-sampling.
S508: according to the corresponding relation of determining, the high fdrequency component of corresponding pixel obtains the synchronization frame high-definition picture in undressed this high-definition picture of pixel value increase with each pixel in the synchronization frame low-resolution image behind the up-sampling.
S509: the async framing low-resolution image is carried out up-sampling process, the size of the async framing low-resolution image after the processing and synchronization frame high-definition picture measure-alike.
S510: the async framing low-resolution image behind the up-sampling is divided into several pixel macroblock, carries out following step S511～S520 for each pixel macroblock.
S511: by the method for estimation, in the synchronization frame high-definition picture of appointment, determine and the immediate pixel macroblock of this pixel macroblock.
Wherein, the synchronization frame high-definition picture of appointment is: be positioned at nearest synchronization frame high-definition picture before this async framing low-resolution image in the first video data, and be arranged in synchronization frame high-definition picture nearest after this async framing low-resolution image at least one.With the immediate pixel macroblock of this pixel macroblock of dividing be: the pixel macroblock of the least absolute error of the pixel value of each pixel and value or least mean-square error value minimum in this pixel macroblock of the pixel value of each pixel that comprises and division.
S512: according to this pixel macroblock of dividing and definite pixel macroblock immediate with it, determine that this pixel macroblock of division is with respect to the motion vector of pixel macroblock immediate with it.
S513: the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in this pixel macroblock of determining to divide according to the motion vector of determining.
S514: according to the corresponding relation of each pixel in each pixel in this pixel macroblock of the pixel value of each pixel in the pixel value of each pixel in this pixel macroblock of dividing, this immediate pixel macroblock, division and this immediate pixel macroblock, determine this pixel macroblock of dividing and the variance rate of this immediate pixel macroblock.
S515: whether the variance rate of judge determining less than setting the variance rate threshold value, if, execution in step S516 then, otherwise execution in step S517.
S516: the pixel value of each pixel increases the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining in this pixel macroblock that will divide.
S517: in the synchronization frame high-definition picture at this immediate pixel macroblock place, determine each neighbor macro block adjacent with this immediate pixel macroblock.
S518: for each pixel in this immediate pixel macroblock, according to the position of this pixel in this immediate pixel macroblock, determine the high fdrequency component of each pixel on the relevant position in each neighbor macro block.
S519: according to the weights of setting for this pixel in this pixel macroblock that connects recently, and be the weights that each pixel on the relevant position is set in each neighbor macro block, weighted average with the high fdrequency component of each pixel on the relevant position in the high fdrequency component of this pixel and each the neighbor macro block redefines the high fdrequency component into this pixel in this immediate pixel macroblock.
S520: the pixel value of each pixel increases the high fdrequency component of each pixel that redefines in the pixel macroblock immediate with it in this pixel macroblock that will divide.
Step S501 shown in Figure 5～S508 is the process of the synchronization frame low-resolution image in the first video data that recovers high frame per second low resolution according to the high-definition picture in the second video data, and step S509～S520 is for recovering the process of the async framing low-resolution image in the first video data according to the synchronization frame high-definition picture that recovers.
So far, the every frame low-resolution image in the first video data (comprising synchronization frame low-resolution image and async framing low-resolution image) is reverted to high-definition picture (the synchronization frame high-definition picture and the async framing high-definition picture that comprise recovery), and then obtained high-resolution the first video data of high frame per second as shown in Figure 3.
The following describes the process of the frame per second of recovering low high-resolution the second video data of frame per second.
In embodiments of the present invention, recovering low high-resolution the second video data of frame per second is to recover according to high-resolution the first video data of high frame per second that has obtained, its core concept is: based on high-resolution the first video data of the high frame per second that obtains, determine in low high-resolution the second video data of frame per second the disappearance two field picture with respect to high-resolution the first video data of this high frame per second, the disappearance two field picture of determining is inserted in the second video data, can recovers the frame per second of the second video data.
Concrete, for the image (high-definition picture that two frames are adjacent) that two frames are adjacent in the second video data, in the first video data after processing (high-resolution the first video data of high frame per second), determine the synchronization frame image (synchronization frame high-definition picture) after two frames corresponding to adjacent with this two frame respectively image are processed, with the async framing image (async framing high-definition picture) after the processing between the synchronization frame image after two frames of determining are processed in the first video data after processing, be defined as the disappearance two field picture to be adjusted between this two frame is adjacent in this second video data the image, determine respectively the parallax of the synchronization frame image after the adjacent image of this two frame and the corresponding processing, according to this two frame of determining adjacent with corresponding processing after the parallax of synchronization frame image, determine the parallax that each disappearance two field picture to be adjusted need to be adjusted, the parallax of adjusting according to each disappearance two field picture needs to be adjusted of determining, adjust disappearance two field picture to be adjusted, the disappearance two field picture after being adjusted.
Because the image in the first video data after processing is high-definition picture, image in the second video data also is high-definition picture, therefore the disappearance two field picture with respect to high-resolution the first video data of high frame per second in the second video data of determining equally also is high-definition picture, and what also namely determine is disappearance vertical frame dimension image in different resolution.
After obtaining the disappearance vertical frame dimension image in different resolution after the adjustment between this two frame is adjacent in the second video data the image, then disappearance vertical frame dimension image in different resolution can be inserted in the adjacent image of this two frame.After all carrying out aforesaid operations for the adjacent high-definition picture of per two frames in low high-resolution the second video data of frame per second, can recover the frame per second of the second video data, obtain high-resolution the second video data of high frame per second.
In said process, according to the parallax of the adjacent image of this two frame of determining with corresponding synchronization frame high-definition picture, determine that the method for the parallax that disappearance two field picture to be adjusted need to be adjusted is specially: the employing formula
Figure BDA0000158998870000171
Determine the parallax that each disappearance two field picture to be adjusted need to be adjusted, wherein, D S1Be the parallax of the synchronization frame high-definition picture that image is corresponding with it the preceding that sorts in the adjacent image of this two frame, D S2For ordering in the adjacent image of this two frame after the parallax of the image synchronization frame high-definition picture corresponding with it, m represents that m frame disappearance two field picture to be adjusted, D are arranged between the adjacent image of this two frame nThe parallax that need to adjust for the n frame disappearance two field picture to be adjusted between the adjacent image of this two frame.
Fig. 6 will hang down the detailed process that high-resolution the second video data of frame per second reverts to high-resolution the second video data of high frame per second for what the embodiment of the invention provided, specifically may further comprise the steps:
S601: for the adjacent high-definition picture of per two frames in the second video data, in high-resolution the first video data of the high frame per second that obtains, determine adjacent with this two frame respectively two frame synchronization vertical frame dimension image in different resolution corresponding to high-definition picture.
The schematic diagram of the disappearance vertical frame dimension image in different resolution between determining of providing of the embodiment of the invention high-definition picture that two frames are adjacent in the second video data is provided Fig. 7.In Fig. 7, high-resolution the first video data of the high frame per second that obtains comprises image A 1, image A 2, image A 3, image A 4, image A 5, comprises image B 1, image B 2 in low high-resolution the second video data of frame per second.As shown in Figure 7, synchronization frame high-definition picture corresponding with the image B 1 in the second video data in high-resolution the first video data of high frame per second is image A 1, and corresponding with the image B 2 in the second video data is image A 5.Need to prove, owing to recovered the resolution of the first video data this moment, therefore resolution and the size of every two field picture in high-resolution the first video data of high frame per second among Fig. 7 are with the resolution of every two field picture in high-resolution the second video data of low frame per second and measure-alike.
S602: each the async framing high-definition picture in high-resolution the first video data of the high frame per second that will obtain between this two frame synchronization vertical frame dimension image in different resolution of determining is defined as the disappearance two field picture to be adjusted between this two frame is adjacent in this second video data the high-definition picture.
Continuation describes as an example of Fig. 7 example, and in Fig. 7, image A 2, image A 3, image A 4 are the disappearance vertical frame dimension image in different resolution to be adjusted between the image B 1 and image B 2 in the second definite video data in high-resolution the first video data of high frame per second.
S603: determine respectively the adjacent high-definition picture and the high-resolution parallax of corresponding synchronization frame of this two frame.
Continuation describes as an example of Fig. 7 example, also, determines the parallax of image B 1 and image A 1, determines the parallax of image B 2 and image A 5, can determine by the method for image calibration and Stereo matching equally when determining parallax.
S604: according to the parallax of synchronization frame high-definition picture corresponding in the adjacent high-definition picture of this two frame of determining and this two frame synchronization vertical frame dimension image in different resolution, determine the parallax that each disappearance two field picture to be adjusted need to be adjusted.
Continuation describes as an example of Fig. 7 example, adopts formula
Figure BDA0000158998870000191
When determining parallax that each disappearance two field picture to be adjusted need to adjust, D S1Be the parallax of image A 1 with image B 1, D S2Be the parallax of image A 5 and image B 2, owing to there are 3 frames disappearance vertical frame dimension to be adjusted image in different resolution (image A 2, image A 3, image A 4), so m is 3, and then image A 2 parallax that needs to adjust is
Figure BDA0000158998870000192
The parallax that image A 3 needs to adjust is The parallax that image A 4 needs to adjust is D s 1 + 3 ( D s 2 - D s 1 ) 4 .
S605: according to the parallax that each disappearance two field picture needs to be adjusted of determining are adjusted, each disappearance two field picture to be adjusted of corresponding adjustment, the disappearance vertical frame dimension image in different resolution after obtaining each and adjusting.
Continuation describes as an example of Fig. 7 example, after having determined parallax that image A 2, image A 3, image A 4 need to adjust, then based on the corresponding adjustment image A 2 of parallax, image A 3, the image A 4 determined, the concrete method of adjusting can be based on definite parallax and disappearance two field picture to be adjusted be carried out corresponding translation, the image A 2 ' after being adjusted, image A 3 ', image A 4 '.
S606: the disappearance vertical frame dimension image in different resolution after will adjusting is inserted in the adjacent high-definition picture of this two frame.
Continuation describes as an example of Fig. 7 example, after obtaining image A 2 ', image A 3 ', image A 4 ', then image A 2 ', image A 3 ', image A 4 ' are sorted according to the clooating sequence in high-resolution the first video data of high frame per second before, image A 2 ', image A 3 ', image A 4 ' after the ordering are inserted into before the image B 1 and image B 2 in the second video data, and then the frame per second of recovering to hang down high-resolution the second video data of frame per second, obtain high-resolution the second video data of high frame per second.
So far, will hang down high-resolution the second video data of frame per second and revert to high-resolution the second video data of high frame per second, and then obtain high-resolution the second video data of high frame per second as shown in Figure 4.
In subsequent step, then can based on obtain be similarly high-resolution the first video data of high frame per second and the second video data carries out subsequent treatment.
Certainly, the method for the resolution of above-mentioned recovery low-resolution video data in the embodiment of the invention, and the method for the frame per second of the low frame-rate video data of recovery is equally applicable to the scene of the above video data of two-way.When having multi-path video data, can adopt the highest video data of its intermediate-resolution to recover the resolution of other video datas, adopt video data that wherein frame per second is the highest to recover the frame per second of other video datas, therefore can reduce equally frame per second and/or the resolution of a few roads video data in the multi-path video data, reach in the display quality that guarantees the 3D video, reduce the purpose of the data volume of 3D video.
Based on above-mentioned same thinking, the embodiment of the invention also provides a kind of video process apparatus, as shown in Figure 8.Fig. 8 is the video process apparatus structural representation that the embodiment of the invention provides, and specifically comprises:
Receiver module 801, be used for receiving the first video data and the second video data, the frame per second of described the first video data is higher than the frame per second of described the second video data, and the resolution of the image in described the first video data is lower than the resolution of the image in described the second video data;
Resolution is recovered module 802, be used for synchronization frame image corresponding to image definite at described the first video data and in described the second video data, based on the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence, according to the synchronization frame image after processing, with the resolution processes of the async framing image in described the first video data be with process after the resolution of synchronization frame image identical, the first video data after obtaining processing;
Frame per second is recovered module 803, be used for according to the first video data after the described processing, determine in described the second video data the disappearance two field picture with respect to the first video data after the described processing, the disappearance two field picture of determining is inserted in described the second video data the second video data after obtaining processing;
Processing module 804 is used for carrying out subsequent treatment based on the first video data after the described processing and the second video data after the described processing.
Described resolution is recovered module 802 and is specifically comprised:
Synchronization frame determining unit 8021 is used for adopting formula
Figure BDA0000158998870000211
Determine in the first video data the synchronization frame image corresponding with image in described the second video data, wherein a is the frame per second of described the first video data, b is the frame per second of described the second video data, M represents the M two field picture in described the second video data, N represents that the N two field picture in described the first video data is synchronization frame image corresponding to M two field picture in described the second video data
Figure BDA0000158998870000212
Expression is in the mode that rounds up pair
Figure BDA0000158998870000213
Round.
Described resolution is recovered module 802 and is specifically comprised:
Synchronization frame recovery unit 8022, be used for the synchronization frame image of described correspondence is carried out the up-sampling processing, image in the size of the synchronization frame image behind the up-sampling and described the second video data measure-alike, determine image in described the second video data and the parallax of the synchronization frame image behind the up-sampling, based on the parallax of determining and the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image behind the up-sampling.
Described synchronization frame recovery unit 8022 specifically is used for, image in described the second video data is carried out down-sampling to be processed, the size of the image behind the down-sampling is measure-alike with described corresponding synchronization frame image, image behind the described down-sampling is carried out up-sampling to be processed, pass through successively the measure-alike of the size of the image behind down-sampling and the up-sampling and the synchronization frame image behind the described up-sampling, determine described image behind down-sampling and the up-sampling and the parallax of the synchronization frame image behind the described up-sampling of passing through successively by the method for image calibration and Stereo matching, as the image in described the second video data of determining and the parallax of the synchronization frame image behind the up-sampling.
Described synchronization frame recovery unit 8022 specifically is used for, determine the pixel value of each pixel in the described image that passes through successively behind down-sampling and the up-sampling, determine the difference of the pixel value of corresponding each pixel in the pixel value of each pixel in the image in described the second video data and the described image that passes through successively behind down-sampling and the up-sampling, high fdrequency component as each pixel in the image in described the second video data, based on the parallax of determining, determine in the image in described the second video data the corresponding relation of each pixel in the synchronization frame image behind each pixel and described up-sampling, according to the corresponding relation of determining, the pixel value of each pixel in the synchronization frame image behind the described up-sampling is increased the high fdrequency component of pixel corresponding in the image in described the second video data, obtain the identical synchronization frame image of resolution of the image in resolution and described the second video data.
Described resolution is recovered module 802 and is specifically comprised:
Async framing recovery unit 8023, being used for that described async framing image is carried out up-sampling processes, the size of the async framing image behind the up-sampling with process after synchronization frame image measure-alike, async framing image behind the up-sampling is divided into several pixel macroblock, for each pixel macroblock of dividing, method by estimation, determine in the synchronization frame image after the processing of appointment and the immediate pixel macroblock of pixel macroblock of dividing, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining; Wherein, synchronization frame image after the processing of described appointment is: be positioned at the synchronization frame image after the processing nearest before the described async framing image in described the first video data, and be arranged in synchronization frame image after the processing nearest after the described async framing image at least one; With the immediate pixel macroblock of pixel macroblock of dividing be: the pixel macroblock of the least absolute error of the pixel value of each pixel and value or least mean-square error value minimum in the pixel value of each pixel that comprises and the pixel macroblock of division.
Described async framing recovery unit 8023 also is used for, the pixel value of each pixel increases before the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining in the pixel macroblock of will divide, according to the pixel macroblock of dividing and definite pixel macroblock immediate with it, determine that the pixel macroblock of division is with respect to the motion vector of pixel macroblock immediate with it, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of determining to divide according to the motion vector of determining, pixel value according to each pixel in the pixel macroblock of dividing, the pixel value of each pixel in this immediate pixel macroblock, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of dividing is determined the pixel macroblock of division and the variance rate of this immediate pixel macroblock; When the pixel value of each pixel increases the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining in the pixel macroblock of will divide, if the variance rate of determining is less than setting the variance rate threshold value, then the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining, if being not less than, the variance rate of determining sets the variance rate threshold value, then in the synchronization frame image after the processing at this immediate pixel macroblock place, determine each neighbor macro block adjacent with this immediate pixel macroblock, for the pixel in this immediate pixel macroblock, according to the position of this pixel in this immediate pixel macroblock, determine the high fdrequency component of each pixel on the relevant position in each neighbor macro block, according to being the weights that this pixel in this immediate pixel macroblock is set, and be the weights that each pixel on the relevant position is set in each neighbor macro block, weighted average with the high fdrequency component of each pixel on the relevant position in the high fdrequency component of this pixel and each the neighbor macro block, redefine the high fdrequency component for this pixel in this immediate pixel macroblock, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel that redefines in the pixel macroblock immediate with it.
Described frame per second is recovered module 803 and specifically is used for, for the adjacent image of two frames in described the second video data, in the first video data after described processing, determine the synchronization frame image after two frames corresponding to adjacent with this two frame respectively image are processed, with the async framing image after the processing between the synchronization frame image after two frames of determining are processed in the first video data after the described processing, be defined as the disappearance two field picture to be adjusted between this two frame is adjacent in described the second video data the image, determine respectively the parallax of the synchronization frame image after the adjacent image of this two frame and the corresponding processing, parallax according to the synchronization frame image after the adjacent image of this two frame of determining and the corresponding processing, determine the parallax that disappearance two field picture to be adjusted need to be adjusted, the parallax of adjusting according to the disappearance two field picture needs to be adjusted of determining, adjust disappearance two field picture to be adjusted, the disappearance two field picture after being adjusted.
Described frame per second is recovered module 803 and specifically is used for, and adopts formula
Figure BDA0000158998870000231
Determine the parallax that each disappearance two field picture to be adjusted need to be adjusted, wherein, D S1Be the parallax of the synchronization frame image after the processing that image is corresponding with it the preceding of sorting in the adjacent image of this two frame, D S2For ordering in the adjacent image of this two frame after the image processing corresponding with it after the parallax of synchronization frame image, m represents that m frame disappearance two field picture to be adjusted, D are arranged between the adjacent image of this two frame nThe parallax that need to adjust for the n frame disappearance two field picture to be adjusted between the adjacent image of this two frame.
Described receiver module 801 specifically is used for, and is received in the first video data and the second video data that the different angles of same collection target collect.
The embodiment of the invention provides a kind of method for processing video frequency and device, the method receives the first video data of high frame per second low resolution, and high-resolution the second video data of low frame per second, and according to the every vertical frame dimension image in different resolution in the second video data, recover the resolution of each image in the first video data, obtain high-resolution the first video data of high frame per second, again according to high-resolution the first video data of high frame per second that obtains, recover the frame per second of the second video data, obtain high-resolution the second video data of high frame per second, carry out subsequent treatment based on high-resolution the first video data of the high frame per second that obtains and high-resolution the second video data of high frame per second.Pass through said method, can be when gathering the 3D video, reduce the resolution of one road video data and the frame per second of another road video data, and when the two-path video data that processing collects, the two-path video data are reverted to high frame per second and high-resolution, therefore can in the display quality that guarantees the 3D video, reduce the data volume of 3D video, save storage resources.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, if of the present invention these are revised and modification belongs within the scope of claim of the present invention and equivalent technologies thereof, then the present invention also is intended to comprise these changes and modification interior.

Claims (20)

1. a method for processing video frequency is characterized in that, comprising:
Receive the first video data and the second video data, the frame per second of described the first video data is higher than the frame per second of described the second video data, and the resolution of the image in described the first video data is lower than the resolution of the image in described the second video data;
Synchronization frame image corresponding to image definite and in described the second video data in described the first video data, based on the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence;
According to the synchronization frame image after processing, be identical with the resolution of synchronization frame image after the processing with the resolution processes of the async framing image in described the first video data, the first video data after obtaining processing;
According to the first video data after the described processing, determine in described the second video data the disappearance two field picture with respect to the first video data after the described processing, the disappearance two field picture of determining is inserted in described the second video data the second video data after obtaining processing;
Carry out subsequent treatment based on the first video data after the described processing and the second video data after the described processing.
2. the method for claim 1 is characterized in that, synchronization frame image corresponding to image definite and in described the second video data in described the first video data specifically comprises:
Adopt formula
Figure FDA0000158998860000011
Determine in the first video data the synchronization frame image corresponding with image in described the second video data, wherein a is the frame per second of described the first video data, b is the frame per second of described the second video data, M represents the M two field picture in described the second video data, N represents that the N two field picture in described the first video data is synchronization frame image corresponding to M two field picture in described the second video data
Figure FDA0000158998860000012
Expression is in the mode that rounds up pair
Figure FDA0000158998860000013
Round.
3. the method for claim 1 is characterized in that, based on the image in described the second video data, is identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence, specifically comprises:
The synchronization frame image of described correspondence is carried out up-sampling process, the image in the size of the synchronization frame image behind the up-sampling and described the second video data measure-alike;
Determine image in described the second video data and the parallax of the synchronization frame image behind the up-sampling, based on the parallax of determining and the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image behind the up-sampling.
4. method as claimed in claim 3 is characterized in that, determines image in described the second video data and the parallax of the synchronization frame image behind the up-sampling, specifically comprises:
Image in described the second video data is carried out down-sampling process, the size of the image behind the down-sampling is measure-alike with described corresponding synchronization frame image;
Image behind the described down-sampling is carried out up-sampling process, pass through successively the measure-alike of the size of the image behind down-sampling and the up-sampling and the synchronization frame image behind the described up-sampling;
Determine described image behind down-sampling and the up-sampling and the parallax of the synchronization frame image behind the described up-sampling of passing through successively by the method for image calibration and Stereo matching, as the image in described the second video data of determining and the parallax of the synchronization frame image behind the up-sampling.
5. method as claimed in claim 4, it is characterized in that, based on the parallax of determining and the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image behind the up-sampling, specifically comprise:
Determine the pixel value of each pixel in the described image that passes through successively behind down-sampling and the up-sampling;
Determine the difference of the pixel value of corresponding each pixel in the pixel value of each pixel in the image in described the second video data and the described image that passes through successively behind down-sampling and the up-sampling, as the high fdrequency component of each pixel in the image in described the second video data;
Based on the parallax of determining, determine in the image in described the second video data the corresponding relation of each pixel in the synchronization frame image behind each pixel and described up-sampling;
According to the corresponding relation of determining, the pixel value of each pixel in the synchronization frame image behind the described up-sampling is increased the high fdrequency component of pixel corresponding in the image in described the second video data, obtain the identical synchronization frame image of resolution of the image in resolution and described the second video.
6. method as claimed in claim 5 is characterized in that, according to the synchronization frame image after processing, be identical with the resolution of synchronization frame image after the processing with the resolution processes of the async framing image in described the first video data, specifically comprises:
Described async framing image is carried out up-sampling processes, the size of the async framing image behind the up-sampling with process after synchronization frame image measure-alike;
Async framing image behind the up-sampling is divided into several pixel macroblock, carries out following step for each pixel macroblock of dividing:
By the method for estimation, determine in the synchronization frame image after the processing of appointment and the immediate pixel macroblock of pixel macroblock of dividing;
Wherein, synchronization frame image after the processing of described appointment is: be positioned at the synchronization frame image after the processing nearest before the described async framing image in described the first video data, and be arranged in synchronization frame image after the processing nearest after the described async framing image at least one; With the immediate pixel macroblock of pixel macroblock of dividing be: the pixel macroblock of the least absolute error of the pixel value of each pixel and value or least mean-square error value minimum in the pixel value of each pixel that comprises and the pixel macroblock of division;
The pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining.
7. method as claimed in claim 6 is characterized in that, before the high fdrequency component of each pixel, described method also comprises in the pixel macroblock immediate with it that the pixel value increase of each pixel in the pixel macroblock of dividing is definite:
According to the pixel macroblock of dividing and definite pixel macroblock immediate with it, determine that the pixel macroblock of division is with respect to the motion vector of pixel macroblock immediate with it;
The corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of determining to divide according to the motion vector of determining;
According to the corresponding relation of each pixel in each pixel in the pixel macroblock of the pixel value of each pixel in the pixel value of each pixel in the pixel macroblock of dividing, this immediate pixel macroblock, division and this immediate pixel macroblock, determine the pixel macroblock of dividing and the variance rate of this immediate pixel macroblock;
The high fdrequency component of each pixel in the pixel macroblock immediate with it that the pixel value increase of each pixel in the pixel macroblock of dividing is definite specifically comprises:
When the variance rate of determining when setting the variance rate threshold value, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining;
When the variance rate of determining is not less than setting variance rate threshold value, in the synchronization frame image after the processing at this immediate pixel macroblock place, determine each neighbor macro block adjacent with this immediate pixel macroblock;
For the pixel in this immediate pixel macroblock, according to the position of this pixel in this immediate pixel macroblock, determine the high fdrequency component of each pixel on the relevant position in each neighbor macro block, according to being the weights that this pixel in this immediate pixel macroblock is set, and be the weights that each pixel on the relevant position is set in each neighbor macro block, weighted average with the high fdrequency component of each pixel on the relevant position in the high fdrequency component of this pixel and each the neighbor macro block redefines the high fdrequency component into this pixel in this immediate pixel macroblock;
The pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel that redefines in the pixel macroblock immediate with it.
8. the method for claim 1 is characterized in that, according to the first video data after the described processing, determines in described the second video data specifically to comprise with respect to the disappearance two field picture of the first video data after the described processing:
For the adjacent image of two frames in described the second video data, in the first video data after described processing, determine the synchronization frame image after two frames corresponding to adjacent with this two frame respectively image are processed;
With the async framing image after the processing between the synchronization frame image after two frames of determining are processed in the first video data after the described processing, be defined as the disappearance two field picture to be adjusted between this two frame is adjacent in described the second video data the image;
Determine respectively the parallax of the synchronization frame image after the adjacent image of this two frame and the corresponding processing;
According to the parallax of the synchronization frame image after the adjacent image of this two frame of determining and the corresponding processing, determine the parallax that disappearance two field picture to be adjusted need to be adjusted;
According to the parallax that the disappearance two field picture needs to be adjusted of determining are adjusted, adjust disappearance two field picture to be adjusted, the disappearance two field picture after being adjusted.
9. method as claimed in claim 8 is characterized in that, according to the parallax of the synchronization frame image after the adjacent image of this two frame of determining and the corresponding processing, determines the parallax that disappearance two field picture to be adjusted need to be adjusted, and specifically comprises:
Adopt formula
Figure FDA0000158998860000051
Determine the parallax that each disappearance two field picture to be adjusted need to be adjusted, wherein, D S1Be the parallax of the synchronization frame image after the processing that image is corresponding with it the preceding of sorting in the adjacent image of this two frame, D S2For ordering in the adjacent image of this two frame after the image processing corresponding with it after the parallax of synchronization frame image, m represents that m frame disappearance two field picture to be adjusted, D are arranged between the adjacent image of this two frame nThe parallax that need to adjust for the n frame disappearance two field picture to be adjusted between the adjacent image of this two frame.
10. the method for claim 1 is characterized in that, receives the first video data and the second video data, specifically comprises:
Be received in the first video data and the second video data that the different angles of same collection target collect.
11. a video process apparatus is characterized in that, comprising:
Receiver module, be used for receiving the first video data and the second video data, the frame per second of described the first video data is higher than the frame per second of described the second video data, and the resolution of the image in described the first video data is lower than the resolution of the image in described the second video data;
Resolution is recovered module, be used for synchronization frame image corresponding to image definite at described the first video data and in described the second video data, based on the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image of correspondence, according to the synchronization frame image after processing, with the resolution processes of the async framing image in described the first video data be with process after the resolution of synchronization frame image identical, the first video data after obtaining processing;
Frame per second is recovered module, be used for according to the first video data after the described processing, determine in described the second video data the disappearance two field picture with respect to the first video data after the described processing, the disappearance two field picture of determining is inserted in described the second video data the second video data after obtaining processing;
Processing module is used for carrying out subsequent treatment based on the first video data after the described processing and the second video data after the described processing.
12. device as claimed in claim 11 is characterized in that, described resolution is recovered module and is specifically comprised:
The synchronization frame determining unit is used for adopting formula
Figure FDA0000158998860000061
Determine in the first video data the synchronization frame image corresponding with image in described the second video data, wherein a is the frame per second of described the first video data, b is the frame per second of described the second video data, M represents the M two field picture in described the second video data, N represents that the N two field picture in described the first video data is synchronization frame image corresponding to M two field picture in described the second video data
Figure FDA0000158998860000062
Expression is in the mode that rounds up pair Round.
13. device as claimed in claim 11 is characterized in that, described resolution is recovered module and is specifically comprised:
The synchronization frame recovery unit, be used for the synchronization frame image of described correspondence is carried out the up-sampling processing, image in the size of the synchronization frame image behind the up-sampling and described the second video data measure-alike, determine image in described the second video data and the parallax of the synchronization frame image behind the up-sampling, based on the parallax of determining and the image in described the second video data, be identical with the resolution of image in described the second video data with the resolution processes of the synchronization frame image behind the up-sampling.
14. device as claimed in claim 13, it is characterized in that, described synchronization frame recovery unit specifically is used for, image in described the second video data is carried out down-sampling to be processed, the size of the image behind the down-sampling is measure-alike with described corresponding synchronization frame image, image behind the described down-sampling is carried out up-sampling to be processed, pass through successively the measure-alike of the size of the image behind down-sampling and the up-sampling and the synchronization frame image behind the described up-sampling, determine described image behind down-sampling and the up-sampling and the parallax of the synchronization frame image behind the described up-sampling of passing through successively by the method for image calibration and Stereo matching, as the image in described the second video data of determining and the parallax of the synchronization frame image behind the up-sampling.
15. device as claimed in claim 14, it is characterized in that, described synchronization frame recovery unit specifically is used for, determine the pixel value of each pixel in the described image that passes through successively behind down-sampling and the up-sampling, determine the difference of the pixel value of corresponding each pixel in the pixel value of each pixel in the image in described the second video data and the described image that passes through successively behind down-sampling and the up-sampling, high fdrequency component as each pixel in the image in described the second video data, based on the parallax of determining, determine in the image in described the second video data the corresponding relation of each pixel in the synchronization frame image behind each pixel and described up-sampling, according to the corresponding relation of determining, the pixel value of each pixel in the synchronization frame image behind the described up-sampling is increased the high fdrequency component of pixel corresponding in the image in described the second video data, obtain the identical synchronization frame image of resolution of the image in resolution and described the second video data.
16. device as claimed in claim 15 is characterized in that, described resolution is recovered module and is specifically comprised:
The async framing recovery unit, being used for that described async framing image is carried out up-sampling processes, the size of the async framing image behind the up-sampling with process after synchronization frame image measure-alike, async framing image behind the up-sampling is divided into several pixel macroblock, for each pixel macroblock of dividing, method by estimation, determine in the synchronization frame image after the processing of appointment and the immediate pixel macroblock of pixel macroblock of dividing, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining; Wherein, synchronization frame image after the processing of described appointment is: be positioned at the synchronization frame image after the processing nearest before the described async framing image in described the first video data, and be arranged in synchronization frame image after the processing nearest after the described async framing image at least one; With the immediate pixel macroblock of pixel macroblock of dividing be: the pixel macroblock of the least absolute error of the pixel value of each pixel and value or least mean-square error value minimum in the pixel value of each pixel that comprises and the pixel macroblock of division.
17. device as claimed in claim 16, it is characterized in that, described async framing recovery unit also is used for, the pixel value of each pixel increases before the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining in the pixel macroblock of will divide, according to the pixel macroblock of dividing and definite pixel macroblock immediate with it, determine that the pixel macroblock of division is with respect to the motion vector of pixel macroblock immediate with it, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of determining to divide according to the motion vector of determining, pixel value according to each pixel in the pixel macroblock of dividing, the pixel value of each pixel in this immediate pixel macroblock, the corresponding relation of each pixel in each pixel and this immediate pixel macroblock in the pixel macroblock of dividing is determined the pixel macroblock of division and the variance rate of this immediate pixel macroblock; When the pixel value of each pixel increases the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining in the pixel macroblock of will divide, if the variance rate of determining is less than setting the variance rate threshold value, then the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel in the pixel macroblock immediate with it of determining, if being not less than, the variance rate of determining sets the variance rate threshold value, then in the synchronization frame image after the processing at this immediate pixel macroblock place, determine each neighbor macro block adjacent with this immediate pixel macroblock, for the pixel in this immediate pixel macroblock, according to the position of this pixel in this immediate pixel macroblock, determine the high fdrequency component of each pixel on the relevant position in each neighbor macro block, according to being the weights that this pixel in this immediate pixel macroblock is set, and be the weights that each pixel on the relevant position is set in each neighbor macro block, weighted average with the high fdrequency component of each pixel on the relevant position in the high fdrequency component of this pixel and each the neighbor macro block, redefine the high fdrequency component for this pixel in this immediate pixel macroblock, the pixel value of each pixel in the pixel macroblock of dividing is increased the high fdrequency component of each pixel that redefines in the pixel macroblock immediate with it.
18. device as claimed in claim 11, it is characterized in that, described frame per second is recovered module and specifically is used for, for the adjacent image of two frames in described the second video data, in the first video data after described processing, determine the synchronization frame image after two frames corresponding to adjacent with this two frame respectively image are processed, with the async framing image after the processing between the synchronization frame image after two frames of determining are processed in the first video data after the described processing, be defined as the disappearance two field picture to be adjusted between this two frame is adjacent in described the second video data the image, determine respectively the parallax of the synchronization frame image after the adjacent image of this two frame and the corresponding processing, parallax according to the synchronization frame image after the adjacent image of this two frame of determining and the corresponding processing, determine the parallax that disappearance two field picture to be adjusted need to be adjusted, the parallax of adjusting according to the disappearance two field picture needs to be adjusted of determining, adjust disappearance two field picture to be adjusted, the disappearance two field picture after being adjusted.
19. device as claimed in claim 18 is characterized in that, described frame per second is recovered module and specifically is used for, and adopts formula
Figure FDA0000158998860000081
Determine the parallax that each disappearance two field picture to be adjusted need to be adjusted, wherein, D S1Be the parallax of the synchronization frame image after the processing that image is corresponding with it the preceding of sorting in the adjacent image of this two frame, D S2For ordering in the adjacent image of this two frame after the image processing corresponding with it after the parallax of synchronization frame image, m represents that m frame disappearance two field picture to be adjusted, D are arranged between the adjacent image of this two frame nThe parallax that need to adjust for the n frame disappearance two field picture to be adjusted between the adjacent image of this two frame.
20. device as claimed in claim 11 is characterized in that, described receiver module specifically is used for, and is received in the first video data and the second video data that the different angles of same collection target collect.
CN201210132682.2A 2012-04-28 2012-04-28 A kind of method for processing video frequency and device Active CN103379351B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210132682.2A CN103379351B (en) 2012-04-28 2012-04-28 A kind of method for processing video frequency and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210132682.2A CN103379351B (en) 2012-04-28 2012-04-28 A kind of method for processing video frequency and device

Publications (2)

Publication Number Publication Date
CN103379351A true CN103379351A (en) 2013-10-30
CN103379351B CN103379351B (en) 2016-03-02

Family

ID=49463836

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210132682.2A Active CN103379351B (en) 2012-04-28 2012-04-28 A kind of method for processing video frequency and device

Country Status (1)

Country Link
CN (1) CN103379351B (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104284158A (en) * 2014-10-23 2015-01-14 南京信必达智能技术有限公司 Event-oriented intelligent camera monitoring method
WO2016029511A1 (en) * 2014-08-27 2016-03-03 深圳市华星光电技术有限公司 Pixel-based method and apparatus for obtaining image with resolution lowered
CN108876716A (en) * 2017-05-11 2018-11-23 Tcl集团股份有限公司 Super resolution ratio reconstruction method and device
CN109819181A (en) * 2019-03-19 2019-05-28 联想(北京)有限公司 Information processing method, device and equipment
CN110827380A (en) * 2019-09-19 2020-02-21 北京铂石空间科技有限公司 Image rendering method and device, electronic equipment and computer readable medium
WO2020073172A1 (en) * 2018-10-08 2020-04-16 Huawei Technologies Co., Ltd. Methods and devices for capturing high-speed and high-definition videos
CN111567030A (en) * 2018-04-23 2020-08-21 Oppo广东移动通信有限公司 Method for automatically switching video capture and playback frame rates
CN111667442A (en) * 2020-05-21 2020-09-15 武汉大学 High-quality high-frame-rate image reconstruction method based on event camera
CN112889268A (en) * 2018-12-21 2021-06-01 理光微电子株式会社 Laser driving device
CN112954251A (en) * 2019-12-10 2021-06-11 RealMe重庆移动通信有限公司 Video processing method, video processing device, storage medium and electronic equipment
CN113015007A (en) * 2021-01-28 2021-06-22 维沃移动通信有限公司 Video frame insertion method and device and electronic equipment
WO2022127565A1 (en) * 2020-12-17 2022-06-23 华为技术有限公司 Video processing method and apparatus, and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1450816A (en) * 2003-04-22 2003-10-22 上海大学 Stereo video stream coder/decoder and stereo video coding/decoding system
CN102036095A (en) * 2010-12-31 2011-04-27 青岛海信信芯科技有限公司 Resolution compensating device and method applied to three-dimensional (3D) image display and 3D television

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1450816A (en) * 2003-04-22 2003-10-22 上海大学 Stereo video stream coder/decoder and stereo video coding/decoding system
CN102036095A (en) * 2010-12-31 2011-04-27 青岛海信信芯科技有限公司 Resolution compensating device and method applied to three-dimensional (3D) image display and 3D television

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016029511A1 (en) * 2014-08-27 2016-03-03 深圳市华星光电技术有限公司 Pixel-based method and apparatus for obtaining image with resolution lowered
CN104284158A (en) * 2014-10-23 2015-01-14 南京信必达智能技术有限公司 Event-oriented intelligent camera monitoring method
CN108876716A (en) * 2017-05-11 2018-11-23 Tcl集团股份有限公司 Super resolution ratio reconstruction method and device
CN111567030A (en) * 2018-04-23 2020-08-21 Oppo广东移动通信有限公司 Method for automatically switching video capture and playback frame rates
CN111567030B (en) * 2018-04-23 2022-01-28 Oppo广东移动通信有限公司 Method for automatically switching video capture and playback frame rates
CN112805996A (en) * 2018-10-08 2021-05-14 华为技术有限公司 Method and equipment for acquiring high-speed high-definition video
US11558549B2 (en) 2018-10-08 2023-01-17 Huawei Technologies Co., Ltd. Methods and devices for capturing high-speed and high-definition videos
CN112805996B (en) * 2018-10-08 2022-04-22 华为技术有限公司 Device and method for generating slow motion video clip
WO2020073172A1 (en) * 2018-10-08 2020-04-16 Huawei Technologies Co., Ltd. Methods and devices for capturing high-speed and high-definition videos
CN112889268A (en) * 2018-12-21 2021-06-01 理光微电子株式会社 Laser driving device
CN112889268B (en) * 2018-12-21 2023-12-15 日清纺微电子有限公司 Laser driving device
CN109819181B (en) * 2019-03-19 2021-02-19 联想(北京)有限公司 Information processing method, device and equipment
CN109819181A (en) * 2019-03-19 2019-05-28 联想(北京)有限公司 Information processing method, device and equipment
CN110827380A (en) * 2019-09-19 2020-02-21 北京铂石空间科技有限公司 Image rendering method and device, electronic equipment and computer readable medium
CN110827380B (en) * 2019-09-19 2023-10-17 北京铂石空间科技有限公司 Image rendering method and device, electronic equipment and computer readable medium
CN112954251A (en) * 2019-12-10 2021-06-11 RealMe重庆移动通信有限公司 Video processing method, video processing device, storage medium and electronic equipment
CN112954251B (en) * 2019-12-10 2023-03-21 RealMe重庆移动通信有限公司 Video processing method, video processing device, storage medium and electronic equipment
CN111667442A (en) * 2020-05-21 2020-09-15 武汉大学 High-quality high-frame-rate image reconstruction method based on event camera
WO2022127565A1 (en) * 2020-12-17 2022-06-23 华为技术有限公司 Video processing method and apparatus, and device
CN113015007A (en) * 2021-01-28 2021-06-22 维沃移动通信有限公司 Video frame insertion method and device and electronic equipment

Also Published As

Publication number Publication date
CN103379351B (en) 2016-03-02

Similar Documents

Publication Publication Date Title
CN103379351B (en) A kind of method for processing video frequency and device
US9525858B2 (en) Depth or disparity map upscaling
CN101472190B (en) Multi-visual angle filming and image processing apparatus and system
CN105100615B (en) A kind of method for previewing of image, device and terminal
CN101291415B (en) Method, apparatus and system for three-dimensional video communication
CN104539929B (en) Stereo-image coding method and code device with motion prediction
CN103238337B (en) Three-dimensional image acquisition system and method
JP2014504462A (en) Video conversion apparatus and display apparatus and method using the same
CN113994366A (en) Multi-stage multi-reference bootstrapping for video super-resolution
CN110650295B (en) Image processing method and device
CN105657247A (en) Secondary exposure photographing method and apparatus for electronic device
CN103260046A (en) Three-dimensional display method and system
WO2016203445A1 (en) MOBILE DEVICE WITH 4 CAMERAS TO TAKE 360ºX 360º STEREOSCOPIC IMAGES AND VIDEOS
CN104185004A (en) Image processing method and image processing system
CN105898283A (en) Synchronous shooting circuit, device and method
CN102742285B (en) Image processing device and image processing method
US20120113221A1 (en) Image processing apparatus and method
DE102019215387A1 (en) CIRCULAR FISH EYE CAMERA ARRAY CORRECTION
EP2415023A1 (en) System and format for encoding data and three-dimensional rendering
CN102186035A (en) Method for displaying screen display information
KR101832407B1 (en) Method and system for communication of stereoscopic three dimensional video information
CN104918037B (en) 3D rendering display device and method and mobile terminal
CN104994365B (en) A kind of method and 2D video three-dimensional methods for obtaining non-key frame depth image
CN107888945A (en) A kind of image transfer method and system
US9532040B2 (en) Virtual viewpoint interval determination sections apparatus and method

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