CN111935417A - Hierarchical video splicing method and device based on multi-scale camera array - Google Patents
Hierarchical video splicing method and device based on multi-scale camera array Download PDFInfo
- Publication number
- CN111935417A CN111935417A CN202010628067.5A CN202010628067A CN111935417A CN 111935417 A CN111935417 A CN 111935417A CN 202010628067 A CN202010628067 A CN 202010628067A CN 111935417 A CN111935417 A CN 111935417A
- Authority
- CN
- China
- Prior art keywords
- video
- level
- local
- global
- resolution
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/265—Mixing
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses a hierarchical video splicing method and a hierarchical video splicing device based on a multi-scale camera array, wherein the method comprises the following steps: acquiring a global video and a plurality of local videos of a scene by a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer; embedding the first-level local video into the global video by using an embedded splicing algorithm to generate a first-level global video; embedding the M-level local video into an M-1-level global video by using an embedded splicing algorithm to generate an M-level global video, wherein M is 2, 3 and 4 … N, and N is the number of levels of a local camera; and finishing the embedding processing of the local videos of all levels to obtain N-level global videos, and outputting the N-level global videos as the global videos of the scene. The method effectively improves the video splicing quality on the premise of considering the splicing speed.
Description
Technical Field
The invention relates to the technical field of video splicing, in particular to a hierarchical video splicing method and device based on a multi-scale camera array.
Background
With the continuous development of computer vision technology, the demand for high-quality image video is increasing, and various acquisition technologies of hundred million-level pixel image video such as single-camera scanning and camera arrays are promoted. The focal lengths of all cameras of a traditional camera array are the same, and after camera pictures are obtained, the camera pictures are spliced one by adopting an optimization strategy. However, the splicing method is not suitable for parallel processing, the splicing speed is low, and accumulated errors are easy to generate.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.
Therefore, one purpose of the invention is to provide a hierarchical video stitching method based on a multi-scale camera array, aiming at the fact that the multi-scale camera array in practical application often has the characteristics of overlapping between adjacent local camera pictures and basically fixed relative positions of the local cameras, sequencing and grading are carried out according to the relative positions of the local cameras, and the global video with high resolution is finally formed by carrying out stitching step by step.
Another objective of the present invention is to provide a hierarchical video stitching apparatus based on multi-scale camera array.
In order to achieve the above object, an embodiment of the present invention provides a hierarchical video stitching method based on a multi-scale camera array, including:
s1, acquiring a global video and a plurality of local videos of a scene through a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer;
s2, embedding the first-level local video into the global video by using an embedded splicing algorithm to generate a first-level global video;
s3, embedding the M-level local video into the M-1 level global video by using an embedded splicing algorithm to generate an M-level global video, wherein M is 2, 3 and 4 … N, and N is the stage number of a local camera;
s4, completing the embedding processing of the local videos of all levels through the S3 to obtain an N-level global video, and outputting the N-level global video as the global video of the scene.
According to the hierarchical video splicing method based on the multi-scale camera array, provided by the embodiment of the invention, aiming at the condition that the field of view of each local camera of the actual multi-scale camera array is overlapped to a certain extent, the high-resolution characteristics between the pictures of adjacent local cameras are matched in a hierarchical splicing mode, and the original matching result is fused, so that the video splicing quality is effectively improved on the premise of considering the splicing speed.
In addition, the hierarchical video stitching method based on the multi-scale camera array according to the above embodiment of the present invention may further have the following additional technical features:
further, in an embodiment of the present invention, before the S1, the method includes: and sequencing and grading according to the relative positions of the local cameras in the multi-scale camera array, wherein the local cameras are divided into N grades, and the local camera of each grade is an adjacent camera of the local camera of the previous grade.
Further, in an embodiment of the present invention, the S2 further includes:
s21, finding out a low-resolution video block corresponding to the first-level local video from the global video collected by the global camera through template matching;
s22, performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm;
s23, performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and S24, fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into the global video acquired by the global camera to generate a first-level global video.
Further, in an embodiment of the present invention, the S3 further includes:
s31, finding out a low-resolution video block corresponding to the M-level local video in the global video collected by the global camera through template matching, and simultaneously taking out a video block B at the same position in the M-1 level global videoM-1；
S32, performing cross-resolution key point matching on the Mth level local video and the corresponding low-resolution video block by using a SURF algorithm;
s33, performing block matching on the M-th level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
s34, using SURF algorithm to combine the M level local video and the video block BM-1Carrying out key point matching under high resolution in the edge area;
and S35, fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth level local video by using a non-uniform twisting algorithm, and embedding the M-th level local video into an M-1 level global video to generate an M-level global video.
Further, in one embodiment of the present invention, each level of local cameras includes at least one local camera.
In order to achieve the above object, an embodiment of another aspect of the present invention provides a hierarchical video stitching apparatus based on a multi-scale camera array, including:
the system comprises an acquisition module, a video acquisition module and a video acquisition module, wherein the acquisition module is used for acquiring a global video and a plurality of local videos of a scene through a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer;
the first splicing module is used for embedding a first-level local video into the global video by utilizing an embedded splicing algorithm to generate a first-level global video;
the second splicing module is used for embedding the M-level local video into the M-1 level global video by using an embedded splicing algorithm to generate an M-level global video, wherein M is 2, 3 and 4 … N, and N is the number of stages of the local camera;
and the output module is used for obtaining N-level global videos after the embedding processing of all levels of local videos is finished, and outputting the N-level global videos as the global videos of the scene.
According to the hierarchical video splicing device based on the multi-scale camera array, the matching of high-resolution key points and the matching of low-resolution key points between pictures of adjacent local cameras are fused in a hierarchical splicing mode, so that the video splicing quality is effectively improved on the premise of considering the splicing speed.
In addition, the hierarchical video stitching apparatus based on the multi-scale camera array according to the above embodiment of the present invention may further have the following additional technical features:
further, in an embodiment of the present invention, the method further includes: a grading module;
the grading module is used for sequencing and grading according to the relative positions of the local cameras in the multi-scale camera array, the local cameras are divided into N grades, and the local camera at each grade is an adjacent camera of the local camera at the previous grade.
Further, in an embodiment of the present invention, the first splicing module is specifically configured to,
finding out a low-resolution video block corresponding to the first-level local video from the global video acquired by the global camera through template matching;
performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into the global video acquired by the global camera to generate a first-level global video.
Further, in one embodiment of the present invention, the second stitched video is specifically for,
finding out a low-resolution video block corresponding to the M-level local video from the global video collected by the global camera through template matching, and simultaneously taking out a video block B at the same position from the M-1 level global videoM-1；
Performing cross-resolution key point matching on the M-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the M-th level local video and a corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth region;
utilizing SURF algorithm to combine the M level local video and the video block BM-1Carrying out key point matching under high resolution in the edge area;
and fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth-level local video by using a non-uniform twisting algorithm, and embedding the M-level local video into an M-1-level global video to generate an M-level global video.
Further, in one embodiment of the present invention, each level of local cameras includes at least one local camera.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flow diagram of a method for hierarchical video stitching based on a multi-scale camera array according to one embodiment of the present invention;
FIG. 2 is a schematic diagram of a manner in which partial cameras of a multi-scale camera array are graded according to one embodiment of the invention;
FIG. 3 is a flow diagram of a hierarchical video stitching algorithm according to one embodiment of the present invention;
fig. 4 is a schematic structural diagram of a hierarchical video stitching apparatus based on a multi-scale camera array according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The following describes a hierarchical video stitching method and apparatus based on a multi-scale camera array according to an embodiment of the present invention with reference to the accompanying drawings.
A proposed hierarchical video stitching method based on a multi-scale camera array according to an embodiment of the present invention will be described first with reference to the accompanying drawings.
FIG. 1 is a flow diagram of a method for hierarchical video stitching based on a multi-scale camera array according to one embodiment of the present invention.
As shown in fig. 1, the hierarchical video stitching method based on the multi-scale camera array comprises the following steps:
step S1, collecting a global video and a plurality of local videos of a scene through a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer.
Further, before S1, the method includes: the local cameras in the multi-scale camera array are sorted and graded according to the relative positions of the local cameras, the local cameras are graded into N grades, and the local camera in each grade is an adjacent camera of the local camera in the previous grade.
As shown in fig. 2, a hierarchical manner of local cameras of a multi-scale camera array is shown according to an embodiment of the present invention, in which 1 global reference camera with a focal length of 8mm and 18 local cameras with a focal length of 50mm are used, and the local cameras are distributed in 3 rows and 6 columns.
The local cameras are divided into 3 levels, 21 levels, 8 2 levels, 8 3 levels. Similarly, other classification methods may be employed as desired.
It is understood that the principle when ranking local cameras is to cover all local cameras with as few levels as possible.
Specifically, as in the above embodiment, the local camera is divided into three stages, the video acquired by the first-stage camera is a first-stage local video, the video acquired by the second-stage camera is a second-stage local video, and the video acquired by the third-stage camera is a third-stage local video.
And step S2, embedding the first-level local video into the global video by using an embedded splicing algorithm to generate a first-level global video.
Further, S2 further includes:
s21, finding out a low-resolution video block corresponding to the first-level local video from the global video collected by the global camera through template matching;
s22, performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm (Speed-Up Robust Features);
s23, performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and S24, fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into the global video acquired by the global camera to generate a first-level global video.
Specifically, a first-level local video is processed to obtain a first-level global video. The local videos of the same level are independent from each other and can be calculated in parallel.
And step S3, embedding the M-level local video into the M-1 level global video by using an embedded stitching algorithm, and generating the M-level global video, wherein M is 2, 3 and 4 … N, and N is the number of levels of the local camera.
And step S4, completing embedding processing of the local videos at all levels through S3 to obtain N-level global videos, and outputting the N-level global videos as global videos of scenes.
Further, S3 further includes:
s31, finding out a low-resolution video block corresponding to the M-level local video from the global video collected by the global camera through template matching, and simultaneously taking out a video block B at the same position from the M-1 level global videoM-1；
S32, performing cross-resolution key point matching on the M-level local video and the corresponding low-resolution video block by using a SURF algorithm;
s33, performing block matching on the M-th level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
s34, using SURF algorithm to combine the M level local video with the video block BM-1Carrying out key point matching under high resolution in the edge area;
and S35, fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth level local video by using a non-uniform twisting algorithm, and embedding the M-th level local video into an M-1 level global video to generate an M-level global video.
When the local videos after the first level are processed, the global videos generated by the previous level are all utilized.
For example, when processing the second-level local video, finding out a low-resolution video block corresponding to the second-level local video from the global video through template matching, and simultaneously taking out a co-located video block B1 from the first-level global video; performing cross-resolution key point matching and block matching on the local video and the corresponding low-resolution video block by using a SURF algorithm and an SM algorithm respectively; performing high-resolution key point matching on the second-level local video and the video block B1 in the edge region by using a SURF algorithm; and fusing matching results, twisting each second-level local video by adopting a non-uniform twisting algorithm, and embedding the twisted second-level local video into the first-level global video to generate a second-level global video.
Further, as with the calculation of the local videos of the first level, the local videos of the same level are independent of each other and can be calculated in parallel.
As shown in fig. 3, it is shown that a first-level local video is processed, a first-level global video is obtained after the first-level local video is processed, a second-level local video is processed according to the first-level global video, and the like until the local videos of all levels are processed, and a global video is obtained.
According to the hierarchical video splicing method based on the multi-scale camera array, provided by the embodiment of the invention, aiming at the condition that the field of view of each local camera of the actual multi-scale camera array is overlapped to a certain extent, the high-resolution characteristics between the pictures of adjacent local cameras are matched in a hierarchical splicing mode, and the original matching result is fused, so that the video splicing quality is effectively improved on the premise of considering the splicing speed.
Next, a hierarchical video stitching apparatus based on a multi-scale camera array according to an embodiment of the present invention will be described with reference to the accompanying drawings.
Fig. 4 is a schematic structural diagram of a hierarchical video stitching apparatus based on a multi-scale camera array according to an embodiment of the present invention.
As shown in fig. 4, the hierarchical video stitching apparatus based on the multi-scale camera array comprises: an acquisition module 100, a first stitching module 200, a second stitching module 300, and an output module 400.
An acquisition module 100 for acquiring a global video and a plurality of local videos of a scene by a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer.
The first stitching module 200 is configured to embed the first-level local video into the global video by using an embedded stitching algorithm, so as to generate a first-level global video.
And a second stitching module 300, configured to embed the M-th level local video into the M-1 level global video by using an embedded stitching algorithm, and generate an M-level global video, where M is 2, 3, and 4 … N, and N is the number of stages of the local camera.
And the output module 400 is configured to obtain an N-level global video after the embedding processing of the local videos at all levels is completed, and output the N-level global video as a global video of a scene.
Further, in an embodiment of the present invention, the method further includes: a grading module;
the grading module is used for sequencing and grading according to the relative positions of the local cameras in the multi-scale camera array, the local cameras are divided into N grades, and the local camera at each grade is an adjacent camera of the local camera at the previous grade.
Further, in one embodiment of the present invention, the first splicing module, in particular for,
finding out a low-resolution video block corresponding to the first-level local video from the global video acquired by the global camera through template matching;
performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into a global video acquired by a global camera to generate a first-level global video.
Further, in one embodiment of the present invention, the second stitched video is specifically for,
finding out a low-resolution video block corresponding to the M-level local video in the global video acquired by the global camera through template matching, and simultaneously taking out a video block B at the same position in the M-1 level global videoM-1；
Performing cross-resolution key point matching on the M-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the M-th level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
using SURF algorithm to combine M level local video with video block BM-1Carrying out key point matching under high resolution in the edge area;
and fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth-level local video by using a non-uniform twisting algorithm, and embedding the M-level local video into an M-1-level global video to generate an M-level global video.
Further, in one embodiment of the present invention, each level of local cameras includes at least one local camera.
It should be noted that the foregoing explanation of the embodiment of the method for stitching a hierarchical video based on a multi-scale camera array is also applicable to the apparatus of the embodiment, and is not repeated here.
According to the hierarchical video splicing device based on the multi-scale camera array, the matching of high-resolution key points and the matching of low-resolution key points between pictures of adjacent local cameras are fused in a hierarchical splicing mode, so that the video splicing quality is effectively improved on the premise of considering the splicing speed.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A hierarchical video stitching method based on a multi-scale camera array is characterized by comprising the following steps:
s1, acquiring a global video and a plurality of local videos of a scene through a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer;
s2, embedding the first-level local video into the global video by using an embedded splicing algorithm to generate a first-level global video;
s3, embedding the M-level local video into the M-1 level global video by using an embedded splicing algorithm to generate an M-level global video, wherein M is 2, 3 and 4 … N, and N is the stage number of a local camera;
s4, completing the embedding processing of the local videos of all levels through the S3 to obtain an N-level global video, and outputting the N-level global video as the global video of the scene.
2. The method for hierarchical video stitching based on multi-scale camera array according to claim 1, wherein before the step S1, the method comprises: and sequencing and grading according to the relative positions of the local cameras in the multi-scale camera array, wherein the local cameras are divided into N grades, and the local camera of each grade is an adjacent camera of the local camera of the previous grade.
3. The method for hierarchical video stitching based on multi-scale camera array according to claim 1, wherein the S2 further comprises:
s21, finding out a low-resolution video block corresponding to the first-level local video from the global video collected by the global camera through template matching;
s22, performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm;
s23, performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and S24, fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into the global video acquired by the global camera to generate a first-level global video.
4. The method for hierarchical video stitching based on multi-scale camera array according to claim 1, wherein the S3 further comprises:
s31, finding out a low-resolution video block corresponding to the M-level local video in the global video collected by the global camera through template matching, and simultaneously taking out a video block B at the same position in the M-1 level global videoM-1；
S32, performing cross-resolution key point matching on the Mth level local video and the corresponding low-resolution video block by using a SURF algorithm;
s33, performing block matching on the M-th level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
s34, using SURF algorithm to combine the M level local video and the video block BM-1Carrying out key point matching under high resolution in the edge area;
and S35, fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth level local video by using a non-uniform twisting algorithm, and embedding the M-th level local video into an M-1 level global video to generate an M-level global video.
5. The method for hierarchical video stitching based on multi-scale camera arrays according to claims 1-4, wherein each level of local cameras comprises at least one local camera.
6. A hierarchical video stitching device based on a multi-scale camera array, comprising:
the system comprises an acquisition module, a video acquisition module and a video acquisition module, wherein the acquisition module is used for acquiring a global video and a plurality of local videos of a scene through a global camera and N levels of local cameras of a multi-scale camera array; wherein N is a positive integer;
the first splicing module is used for embedding a first-level local video into the global video by utilizing an embedded splicing algorithm to generate a first-level global video;
the second splicing module is used for embedding the M-level local video into the M-1 level global video by using an embedded splicing algorithm to generate an M-level global video, wherein M is 2, 3 and 4 … N, and N is the number of stages of the local camera;
and the output module is used for obtaining N-level global videos after the embedding processing of all levels of local videos is finished, and outputting the N-level global videos as the global videos of the scene.
7. The hierarchical video stitching device based on the multi-scale camera array according to claim 6, further comprising: a grading module;
the grading module is used for sequencing and grading according to the relative positions of the local cameras in the multi-scale camera array, the local cameras are divided into N grades, and the local camera at each grade is an adjacent camera of the local camera at the previous grade.
8. The hierarchical video stitching device based on a multi-scale camera array according to claim 6, characterized in that the first stitching module is specifically configured to,
finding out a low-resolution video block corresponding to the first-level local video from the global video acquired by the global camera through template matching;
performing cross-resolution key point matching on the first-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the first-level local video and the corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth area;
and fusing the cross-resolution key point matching result and the block matching result, twisting each first-level local video by using a non-uniform twisting algorithm, and embedding the first-level local video into the global video acquired by the global camera to generate a first-level global video.
9. The hierarchical video stitching device based on the multi-scale camera array according to claim 6, characterized in that the second stitched video is specifically used for,
finding out a low-resolution video block corresponding to the M-level local video from the global video collected by the global camera through template matching, and simultaneously taking out a video block B at the same position from the M-1 level global videoM-1；
Performing cross-resolution key point matching on the M-level local video and the corresponding low-resolution video block by using a SURF algorithm;
performing block matching on the M-th level local video and a corresponding low-resolution video block by adopting a stereo matching SM algorithm, and supplementing key points of a smooth region;
utilizing SURF algorithm to combine the M level local video and the video block BM-1Carrying out key point matching under high resolution in the edge area;
and fusing the cross-resolution key point matching result, the block matching result and the key point matching result under high resolution, twisting each Mth-level local video by using a non-uniform twisting algorithm, and embedding the M-level local video into an M-1-level global video to generate an M-level global video.
10. The hierarchical video stitching device based on the multi-scale camera array according to the claims 6-9, characterized in that each level of the local cameras comprises at least one local camera.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010628067.5A CN111935417B (en) | 2020-07-02 | 2020-07-02 | Hierarchical video splicing method and device based on multi-scale camera array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010628067.5A CN111935417B (en) | 2020-07-02 | 2020-07-02 | Hierarchical video splicing method and device based on multi-scale camera array |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111935417A true CN111935417A (en) | 2020-11-13 |
| CN111935417B CN111935417B (en) | 2021-12-14 |
Family
ID=73317758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010628067.5A Active CN111935417B (en) | 2020-07-02 | 2020-07-02 | Hierarchical video splicing method and device based on multi-scale camera array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111935417B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381894A (en) * | 2021-01-15 | 2021-02-19 | 清华大学 | Adaptive light field imaging calibration method, device and storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040201672A1 (en) * | 2003-04-11 | 2004-10-14 | Sridhar Varadarajan | System and method for warning drivers based on road curvature |
| US20100158311A1 (en) * | 2007-08-31 | 2010-06-24 | Fujitsu Limited | Electronic watermark embedding device, electronic watermark detecting device, and programs therefor |
| CN101976435A (en) * | 2010-10-07 | 2011-02-16 | 西安电子科技大学 | Combination learning super-resolution method based on dual constraint |
| CN103345763A (en) * | 2013-06-25 | 2013-10-09 | 西安理工大学 | Method for calculating motion attention based on multiscale variable-block |
| CN107133921A (en) * | 2016-02-26 | 2017-09-05 | 北京大学 | The image super-resolution rebuilding method and system being embedded in based on multi-level neighborhood |
| CN107659774A (en) * | 2017-09-30 | 2018-02-02 | 深圳市未来媒体技术研究院 | A kind of video imaging system and method for processing video frequency based on multiple dimensioned camera array |
| CN107959805A (en) * | 2017-12-04 | 2018-04-24 | 深圳市未来媒体技术研究院 | Light field video imaging system and method for processing video frequency based on Hybrid camera array |
| CN108038893A (en) * | 2017-12-06 | 2018-05-15 | 深圳市未来媒体技术研究院 | A kind of generation method of 1,000,000,000 pixel videos based on Hybrid camera array |
| CN108156383A (en) * | 2017-12-29 | 2018-06-12 | 清华大学 | 1,000,000,000 pixel video acquisition method of high dynamic and device based on camera array |
| CN109919013A (en) * | 2019-01-28 | 2019-06-21 | 浙江英索人工智能科技有限公司 | Method for detecting human face and device in video image based on deep learning |
| CN110494891A (en) * | 2017-04-14 | 2019-11-22 | 文塔纳医疗系统公司 | It local registration for splicing based on piecemeal and global places |
| CN110503606A (en) * | 2019-08-29 | 2019-11-26 | 广州大学 | A method of improving face clarity |
| CN111028153A (en) * | 2019-12-09 | 2020-04-17 | 南京理工大学 | Image processing and neural network training method and device and computer equipment |
-
2020
- 2020-07-02 CN CN202010628067.5A patent/CN111935417B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040201672A1 (en) * | 2003-04-11 | 2004-10-14 | Sridhar Varadarajan | System and method for warning drivers based on road curvature |
| US20100158311A1 (en) * | 2007-08-31 | 2010-06-24 | Fujitsu Limited | Electronic watermark embedding device, electronic watermark detecting device, and programs therefor |
| CN101976435A (en) * | 2010-10-07 | 2011-02-16 | 西安电子科技大学 | Combination learning super-resolution method based on dual constraint |
| CN103345763A (en) * | 2013-06-25 | 2013-10-09 | 西安理工大学 | Method for calculating motion attention based on multiscale variable-block |
| CN107133921A (en) * | 2016-02-26 | 2017-09-05 | 北京大学 | The image super-resolution rebuilding method and system being embedded in based on multi-level neighborhood |
| CN110494891A (en) * | 2017-04-14 | 2019-11-22 | 文塔纳医疗系统公司 | It local registration for splicing based on piecemeal and global places |
| CN107659774A (en) * | 2017-09-30 | 2018-02-02 | 深圳市未来媒体技术研究院 | A kind of video imaging system and method for processing video frequency based on multiple dimensioned camera array |
| CN107959805A (en) * | 2017-12-04 | 2018-04-24 | 深圳市未来媒体技术研究院 | Light field video imaging system and method for processing video frequency based on Hybrid camera array |
| CN108038893A (en) * | 2017-12-06 | 2018-05-15 | 深圳市未来媒体技术研究院 | A kind of generation method of 1,000,000,000 pixel videos based on Hybrid camera array |
| CN108156383A (en) * | 2017-12-29 | 2018-06-12 | 清华大学 | 1,000,000,000 pixel video acquisition method of high dynamic and device based on camera array |
| CN109919013A (en) * | 2019-01-28 | 2019-06-21 | 浙江英索人工智能科技有限公司 | Method for detecting human face and device in video image based on deep learning |
| CN110503606A (en) * | 2019-08-29 | 2019-11-26 | 广州大学 | A method of improving face clarity |
| CN111028153A (en) * | 2019-12-09 | 2020-04-17 | 南京理工大学 | Image processing and neural network training method and device and computer equipment |
Non-Patent Citations (1)
| Title |
|---|
| 刘金涛等: "《基于多尺度分析的多源视频拼接方法》", 《计算机工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381894A (en) * | 2021-01-15 | 2021-02-19 | 清华大学 | Adaptive light field imaging calibration method, device and storage medium |
| CN112381894B (en) * | 2021-01-15 | 2021-04-20 | 清华大学 | Adaptive light field imaging calibration method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111935417B (en) | 2021-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109696788B (en) | Quick automatic focusing method based on display panel | |
| CN112365404B (en) | Contact net panoramic image splicing method, system and equipment based on multiple cameras | |
| CN101394573B (en) | Panoramagram generation method and system based on characteristic matching | |
| JP5549345B2 (en) | Sky detection apparatus and method used in image acquisition apparatus | |
| CN109961399A (en) | Optimal stitching line method for searching based on Image distance transform | |
| CN111445389A (en) | Wide-view-angle rapid splicing method for high-resolution images | |
| CN102932605A (en) | Method for selecting camera combination in visual perception network | |
| CN103902953B (en) | A kind of screen detecting system and method | |
| CN111340701A (en) | Circuit board image splicing method for screening matching points based on clustering method | |
| CN110544202B (en) | Parallax image splicing method and system based on template matching and feature clustering | |
| CN111935417B (en) | Hierarchical video splicing method and device based on multi-scale camera array | |
| JP7387261B2 (en) | Information processing device, information processing method and program | |
| CN111783729A (en) | Video classification method, device, equipment and storage medium | |
| CN113516116B (en) | Text detection method, system and medium suitable for complex natural scene | |
| CN1606345A (en) | Apparatus and method for performing inverse telecine process | |
| CN104867125A (en) | Image obtaining method and image obtaining device | |
| CN108596837A (en) | Image split-joint method, device, equipment and computer media | |
| CN111476714B (en) | Cross-scale image splicing method and device based on PSV neural network | |
| CN108156383B (en) | High-dynamic billion pixel video acquisition method and device based on camera array | |
| CN106331848A (en) | Panoramic video identification method and device, and video playing method and device | |
| CN115578260B (en) | Attention method and system for directional decoupling of image super-resolution | |
| CN117496319A (en) | All-weather target detection system and method based on unregistered bimodal image fusion | |
| CN110728661A (en) | Image distortion evaluation network training method and device based on randomly generated samples | |
| US9380285B2 (en) | Stereo image processing method, stereo image processing device and display device | |
| JP2018124740A (en) | Image retrieval system, image retrieval method and image retrieval program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |