CN106485781A - Three-dimensional scene construction method based on live video stream and its system - Google Patents
Three-dimensional scene construction method based on live video stream and its system Download PDFInfo
- Publication number
- CN106485781A CN106485781A CN201610872000.XA CN201610872000A CN106485781A CN 106485781 A CN106485781 A CN 106485781A CN 201610872000 A CN201610872000 A CN 201610872000A CN 106485781 A CN106485781 A CN 106485781A
- Authority
- CN
- China
- Prior art keywords
- photographic head
- video stream
- live video
- dimensional
- image
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Studio Devices (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of three-dimensional scene construction method based on live video stream, comprise the steps：Setting standard card cage；Under standard card cage, lay and correct the pose of some photographic head, so that the shooting picture ordered arrangement in each photographic head, and splice；Obtain the image that photographic head shoots in same time frame；The characteristic frame chosen in described image carries out panoramic mosaic；Verify spliced effect；Integrate panoramic view data source, and be switched to video source；Issue video source.The invention also discloses a kind of three-dimensional scenic constructing system based on live video stream.Using the three-dimensional scene construction method based on live video stream of the present invention, dynamic scene can be carried out with three-dimensional reduction, have the characteristics that real-time, low cost, process cycle be short, reduction is true to nature.
Description
Technical field
The invention belongs to virtual emulation, mixed display, geographical perception dimensional Modeling Technology field are and in particular to one kind is based on
The three-dimensional scene construction method of live video stream.
Background technology
With the continuous improvement of the level of informatization, the space-time demand that people are constituted to time dimension and Spatial Dimension is got over
Come bigger because it can help people to capture the information that three-dimensional real-time image cannot present, contribute to people realize decision-making,
The targets such as monitoring.The structure of three-dimensional scenic, generally has two kinds at present：First kind of way is sketched out using 3 d modeling software
Come, such as the instrument such as 3DMax and Mayan；The second way passes through pan-shot, using image processing techniquess, such as AutoPano etc..
For the mode of the first three-dimensional software modeling, there is problems with：
(1) it is only applicable to the modeling that object source data amount is big and dimension is abundant, there is limitation；
(2) substantial amounts of artificial, time, data acquisition cost are needed in processing procedure；
(3) three-dimensional scenic it is difficult to a persistent period dimension can only be built by the three dimensions based on a time point
Reduction.
For second although having the advantages that low cost, cycle is short by way of pan-shot, but still exist with
Lower problem：
(1) during building three-dimensional scenic, it is by strange land data acquisition, therefore there is large scene reduction and scene becomes
The problems such as shape distortion, is serious, therefore limits the application of the method；
(2) three dimensions of a persistent period dimension cannot be reduced.
Both the above mode be all belonging to still image modeling it is impossible to for user provide in a period of time dynamic
State three-dimensional scenic.
Content of the invention
In order to solve the above problems, the first object of the present invention is：A kind of three-dimensional scenic based on live video stream is provided
Construction method, can carry out the three-dimensional reduction of video flowing to dynamic scene, have that real-time, low cost, process cycle be short, reduction
Feature true to nature.
For achieving the above object, the present invention is achieved by technical scheme below：
Three-dimensional scene construction method based on live video stream of the present invention, comprises the steps：
Setting standard card cage；
Under standard card cage, lay and correct the pose of some photographic head, so that the shooting picture in each photographic head has
Sequence arranges, and splices；
Obtain the image that photographic head shoots in same time frame；
The characteristic frame chosen in described image carries out panoramic mosaic；
Verify spliced effect；
Integrate panoramic view data source, and be switched to video source；
Issue three-dimensional scenic.
Further, lay and correct some photographic head step in, the ken of described photographic head covers whole scene,
And there is the overlap of 15-25% in the ken of adjacent camera.
Further, lay and correct some photographic head step in, some photographic head are uniformly distributed in scene
Center is on the circumference in the center of circle, and the depth of field gap of each photographic head is less than the 20% of described radius of a circle.
Further, the resolution of described photographic head is not less than 1080p, visual angle is less than 110 °.
Further, the step that the described fisrt feature frame chosen in described image carries out panoramic mosaic, specifically includes：
According to the orientation of photographic head, by the image of same time frame according to counterclockwise or be ranked up clockwise；
By the image zooming-out feature between adjacent camera；
Feature after extracting is carried out coupling and parameter calculates；
Image co-registration.
Further, the step picture between adjacent cameras being extracted feature, specifically：
Using SIFT algorithm and SURF algorithm as algorithm, specifically include step, the characteristic point of metric space extremum extracting
The step that the step of positioning, the step of direction assignment and feature point description symbol calculate；
Wherein, the step of described metric space extremum extracting, specifically：
Gaussian pyramid is carried out using SIFT algorithm to image and DOG decomposes, to set up the dimensional space of image；
The step of described positioning feature point, specifically：
In the field of DOG pyramidal 3 × 3 × 3, maximum solution and sub-pixel positioning are carried out to characteristic point；
The step of described direction assignment, specifically：
Each characteristic point for extracting gives a principal direction according to field Gradient distribution；
The step that described feature point description symbol calculates, specifically：
The coordinate of each characteristic point, yardstick and principal direction are quantified, forms the feature descriptor of 128 dimensions, and right
The vector of described 128 dimensions is normalized.
Further, the feature after extracting is carried out mating the step with parameter estimation, specifically：With two characteristic points
, as the similarity criteria of Feature Points Matching, Euclidean distance is minimum and is less than certain threshold value for Euclidean distance between characteristic vector
Two characteristic points be matching double points.
In order to solve the above problems, the second object of the present invention is：A kind of three-dimensional scenic based on live video stream is provided
Constructing system, the three-dimensional reduction that this system can carry out video flowing for dynamic scene is provided with niche plinth, and passes through this system
The three-dimensional scenic of reduction has the characteristics that real-time, low cost, process cycle be short, reduction is true to nature.
For achieving the above object, the present invention is achieved by technical scheme below：
Three-dimensional scenic constructing system based on live video stream of the present invention, including：
Setup module, for arranging standard card cage；
Lay and correction module, for, under standard card cage, laying and correcting the pose of some photographic head, so that respectively taking the photograph
As the shooting picture ordered arrangement in head, and splice；
Acquisition module, for obtaining the image that photographic head shoots in same time frame；
Choose module, carry out panoramic mosaic for choosing the characteristic frame in described image；
Correction verification module, for verifying spliced effect；
Integrate and handover module, for integrating panoramic view data source, and be switched to video source；
Release module, for issuing three-dimensional scenic.
Further, in described laying with correction module, the ken of described photographic head covers whole scene, and adjacent
There is the overlap of 15-25% in the ken of photographic head.
Further, in described laying with correction module, some photographic head are uniformly distributed in the center of scene as circle
On the circumference of the heart, and the depth of field gap of each photographic head is less than the 20% of described radius of a circle.
Compared with prior art, the invention has the beneficial effects as follows：
Using the three-dimensional scene construction method based on live video stream of the present invention, by obtaining photographic head shooting
Before picture, process is carried out in just laying and correction to some camera positions, thus ensure that the picture ordered arrangement shot,
The plenty of time in later stage is avoided to be processed, picture splicing is it is ensured that fidelity during follow-up splicing again simultaneously.Then basis regards
The video stream data of frequency photographic head, quick calculating, simulate real-time three-dimensional spatial scene, and utilize data storage technology, permissible
This scene is carried out with time and the reduction of space four dimensions, low cost, place remote, the four-dimension persistently followed the tracks of can be provided the user
Spatial Information Service.
Of the present invention based in the three-dimensional scene construction method of live video stream using process to video flowing, and existing
With the presence of the difference simply static picture etc. being processed in technology substantially, it is based on three-dimensional video and fourth dimension time
Combination, the video a time period is quickly calculated according to chronological order, simulates three-dimensional scenic, solve existing
In technology for four-dimensional scene reduction demand, but the difficult problem that cannot simulate, is that modeling on the four-dimension for the scene opens
New era.
Brief description
Below in conjunction with the accompanying drawings the specific embodiment of the present invention is described in further detail, wherein：
Fig. 1 is the schematic flow sheet of the three-dimensional modeling method of reducing giving live video stream of the present invention；
Fig. 2 is the theory diagram of the three-dimensional modeling also original system giving live video stream of the present invention.
1：Setup module 2：Lay and correction module 3：Acquisition module 4：Choose module
5：Correction verification module 6：Integrate and handover module 7：Release module
Specific embodiment
Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are illustrated it will be appreciated that preferred reality described herein
Apply example to be merely to illustrate and explain the present invention, be not intended to limit the present invention.
As shown in figure 1, the three-dimensional scene construction method based on live video stream of the present invention, comprise the steps：
S01：Setting standard card cage；
The purpose of this standard card cage setting is to provide standard foundation for laying photographic head, so that subsequently to three-dimensional scenic also
Former carry out Initialize installation.Concrete then be geometry according to the actual scene scene framework to determine multiple standards, in reality
During border uses, then choose suitable framework as standard card cage.
S02：Under standard card cage, lay and correct the pose of some photographic head, so that the shooting picture in each photographic head
Face ordered arrangement, and splice；
Wherein, the ken of described photographic head covers whole scene, to find a view completely, the ken of adjacent camera is deposited simultaneously
Process it is ensured that facilitating in follow-up splicing in the overlap of 15-25%；
In addition, in the laying of described photographic head, some photographic head are uniformly distributed in the circumference with the center of scene as the center of circle
On, and the depth of field gap of each photographic head is less than the 20% of described radius of a circle, and purpose is just to ensure that each photographic head is clapped
The image taken the photograph can be preferably that follow-up splicing reduces the environment processing, it is to avoid the problems such as distortion.Wherein, for the shooting selecting
Head, its resolution otherwise less than 1080p, visual angle be less than 110 ° it is intended to improve shooting effect.
After above laying, also need the depth of field of the pose, the central point of photographic head and photographic head of inspection and correction photographic head
Whether meet preset standard.
S03：Obtain the image that photographic head shoots in same time frame；
Guarantee that photographic head is to carry out in shooting process simultaneously, be not in the appearance of the problem of dtmf distortion DTMF that time difference leads to.
S04：The characteristic frame chosen in described image carries out panoramic mosaic；
This step specifically includes again：
S041：According to the orientation of photographic head, by the image of same time frame according to counterclockwise or be ranked up clockwise；
S042：By the image zooming-out feature between adjacent camera；
Adopt SIFT algorithm and SURF algorithm as algorithm in this step, specifically include the step of metric space extremum extracting
Suddenly, the step that the step of positioning feature point, the step of direction assignment and feature point description symbol calculate；
Wherein, the step of described metric space extremum extracting, specifically：
Gaussian pyramid is carried out using SIFT algorithm to image and DOG decomposes, to set up the dimensional space of image；
The step of described positioning feature point, specifically：
In the field of DOG pyramidal 3 × 3 × 3, maximum solution and sub-pixel positioning are carried out to characteristic point；
The step of described direction assignment, specifically：
Each characteristic point for extracting gives a principal direction according to field Gradient distribution；
The step that described feature point description symbol calculates, specifically：
The coordinate of each characteristic point, yardstick and principal direction are quantified, forms the feature descriptor of 128 dimensions, and right
The vector of described 128 dimensions is normalized, to remove the impact that illumination variation is brought.
S043：Feature after extracting is carried out coupling and parameter calculates；
Using the Euclidean distance between the characteristic vector of two characteristic points as the similarity criteria of Feature Points Matching, choose Europe
Formula is matching double points apart from minimum and less than certain threshold value two characteristic points.In this step, between to different images
Characteristic point when being mated, need characteristic vector is carried out with the adjacent search of nearest k-d tree.
S044：Image co-registration；
This step includes many flat rubber beltings blending algorithm of spherical projection method or tower algorithm；
Wherein, described spherical projection method is specifically：
Image unification is projected to image blend process is carried out on same sphere；
Many flat rubber beltings blending algorithm of described tower algorithm is specifically：
Image is decomposed in different frequency bands；
Take after different Tw values are weighted interpolation processing respectively and merged, with realize simultaneously smoother transition with
And preferable high frequency detail reserve capability.
S05：Verify spliced effect；
S06：Integrate panoramic view data source, and be switched to video source；
S07：Issue three-dimensional scenic.
As shown in Fig. 2 the three-dimensional scenic constructing system based on live video stream of the present invention, including：
Setup module 1, for arranging standard card cage；
Lay and correction module 2, for, under standard card cage, laying and correcting the pose of some photographic head, so that respectively
Shooting picture ordered arrangement in photographic head, and splice；
Wherein, the ken of described photographic head covers whole scene, and the ken of adjacent camera has the weight of 15-25%
Folded, optimum selection is then 20% overlap, for the process that follow-up splicing is more convenient.Meanwhile, some photographic head are uniform
It is distributed on the circumference with the center of scene as the center of circle, and the depth of field gap of each photographic head is less than described radius of a circle
20% it is intended to ensure that the image that each photographic head shoots can be preferably that follow-up splicing reduces the environment processing, it is to avoid distortion etc. is asked
Topic.
Acquisition module 3, for obtaining the image that photographic head shoots in same time frame；
Choose module 4, carry out panoramic mosaic for choosing the characteristic frame in described image；
Correction verification module 5, for verifying spliced effect；
Integrate and handover module 6, for integrating panoramic view data source, and be switched to video source；
Release module 7, for issuing three-dimensional scenic.
Three-dimensional scene construction method based on live video stream of the present invention and system, by way of shooting, and
And the pose of photographic head effectively will be adjusted before shooting, it is that the splicing of subsequent figures picture is greatly reduced workload, and
Also ensure that the effectiveness obtaining image.Meanwhile, in conjunction with the video flowing that photographic head is caught, and non-static picture it is ensured that
Three dimensions video and the 4th temporal combination, significantly meet the demand to four-dimensional scenario reduction for the every profession and trade, are also existing
There is technology to start four-dimensional image and start new era of knowing clearly.
The above, be only presently preferred embodiments of the present invention, not the present invention is made with any pro forma restriction, therefore
Every any modification without departing from technical solution of the present invention content, according to the technical spirit of the present invention, above example made,
Equivalent variations and modification, all still fall within the range of technical solution of the present invention.
Claims (10)
1. a kind of three-dimensional scene construction method based on live video stream is it is characterised in that comprise the steps：
Setting standard card cage；
Under standard card cage, lay and correct the pose of some photographic head, so that the shooting picture in each photographic head is arranged in order
Row, and splice；
Obtain the image that photographic head shoots in same time frame；
The characteristic frame chosen in described image carries out panoramic mosaic；
Verify spliced effect；
Integrate panoramic view data source, and be switched to video source；
Issue three-dimensional scenic.
2. the three-dimensional scene construction method based on live video stream according to claim 1 it is characterised in that：
Lay and correct some photographic head step in, the ken of described photographic head covers whole scene, and adjacent shooting
There is the overlap of 15-25% in the ken of head.
3. the three-dimensional scene construction method based on live video stream according to claim 1 and 2 it is characterised in that：
Lay and correct some photographic head step in, some photographic head are uniformly distributed in the circle with the center of scene as the center of circle
Zhou Shang, and the depth of field gap of each photographic head is less than the 20% of described radius of a circle.
4. the three-dimensional scene construction method based on live video stream according to claim 1 it is characterised in that：
The resolution of described photographic head is not less than 1080p, visual angle is less than 110 °.
5. the three-dimensional scene construction method based on live video stream according to claim 1 it is characterised in that：
The step that characteristic frame in described selection described image carries out panoramic mosaic, specifically includes：
According to the orientation of photographic head, by the image of same time frame according to counterclockwise or be ranked up clockwise；
By the image zooming-out feature between adjacent camera；
Feature after extracting is carried out coupling and parameter calculates；
Image co-registration.
6. the three-dimensional scene construction method based on live video stream according to claim 5 it is characterised in that：
The step that picture between adjacent cameras is extracted feature, specifically：
Using SIFT algorithm and SURF algorithm as algorithm, specifically include step, the positioning feature point of metric space extremum extracting
Step, the step of direction assignment and feature point description symbol calculate step；
Wherein, the step of described metric space extremum extracting, specifically：
Gaussian pyramid is carried out using SIFT algorithm to image and DOG decomposes, to set up the dimensional space of image；
The step of described positioning feature point, specifically：
In the field of DOG pyramidal 3 × 3 × 3, maximum solution and sub-pixel positioning are carried out to characteristic point；
The step of described direction assignment, specifically：
Each characteristic point for extracting gives a principal direction according to field Gradient distribution；
The step that described feature point description symbol calculates, specifically：
The coordinate of each characteristic point, yardstick and principal direction are quantified, forms the feature descriptor of 128 dimensions, and to described
The vector of 128 dimensions is normalized.
7. the three-dimensional scene construction method based on live video stream according to claim 5 it is characterised in that：
Feature after extracting is carried out mating the step with parameter estimation, specifically：
Using the Euclidean distance between the characteristic vector of two characteristic points as the similarity criteria of Feature Points Matching, Euclidean distance is
Little and less than certain threshold value two characteristic points are matching double points.
8. a kind of three-dimensional scenic constructing system based on live video stream is it is characterised in that include：
Setup module, for arranging standard card cage；
Lay and correction module, for, under standard card cage, laying and correcting the pose of some photographic head, so that each photographic head
In shooting picture ordered arrangement, and splice；
Acquisition module, for obtaining the image that photographic head shoots in same time frame；
Choose module, carry out panoramic mosaic for choosing the characteristic frame in described image；
Correction verification module, for verifying spliced effect；
Integrate and handover module, for integrating panoramic view data source, and be switched to video source；
Release module, for issuing three-dimensional scenic.
9. the three-dimensional scenic constructing system based on live video stream according to claim 8 it is characterised in that：
In described laying with correction module, the ken of described photographic head covers whole scene, and the ken of adjacent camera
There is the overlap of 15-25%.
10. the three-dimensional scenic constructing system based on live video stream according to claim 8 or claim 9 it is characterised in that：
In described laying with correction module, some photographic head are uniformly distributed on the circumference with the center of scene as the center of circle, and
And the depth of field gap of each photographic head is less than the 20% of described radius of a circle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610872000.XA CN106485781A (en) | 2016-09-30 | 2016-09-30 | Three-dimensional scene construction method based on live video stream and its system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610872000.XA CN106485781A (en) | 2016-09-30 | 2016-09-30 | Three-dimensional scene construction method based on live video stream and its system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106485781A true CN106485781A (en) | 2017-03-08 |
Family
ID=58268356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610872000.XA Pending CN106485781A (en) | 2016-09-30 | 2016-09-30 | Three-dimensional scene construction method based on live video stream and its system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106485781A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107901424A (en) * | 2017-12-15 | 2018-04-13 | 北京中睿华信信息技术有限公司 | A kind of Image Acquisition modeling |
| CN109147798A (en) * | 2018-07-27 | 2019-01-04 | 北京三快在线科技有限公司 | Audio recognition method, device, electronic equipment and readable storage medium storing program for executing |
| CN109978987A (en) * | 2017-12-28 | 2019-07-05 | 周秦娜 | A kind of control method, apparatus and system constructing panorama based on multiple depth cameras |
| CN112650422A (en) * | 2020-12-17 | 2021-04-13 | 咪咕文化科技有限公司 | AR interaction method and device of equipment, electronic equipment and storage medium |
| CN115294748A (en) * | 2022-09-08 | 2022-11-04 | 广东中科凯泽信息科技有限公司 | Fixed target disappearance early warning method based on visual data analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103236160A (en) * | 2013-04-07 | 2013-08-07 | 水木路拓科技(北京)有限公司 | Road network traffic condition monitoring system based on video image processing technology |
| US20150054913A1 (en) * | 2013-08-21 | 2015-02-26 | Jaunt Inc. | Image stitching |
| CN104767975A (en) * | 2015-04-10 | 2015-07-08 | 武汉市测绘研究院 | Method for achieving interactive panoramic video stream map |
| CN104778675A (en) * | 2015-04-28 | 2015-07-15 | 中国矿业大学 | Coal mining fully mechanized excavation face dynamic video image fusion method |
| CN105262949A (en) * | 2015-10-15 | 2016-01-20 | 浙江卓锐科技股份有限公司 | Multifunctional panorama video real-time splicing method |
| US20160088285A1 (en) * | 2014-09-22 | 2016-03-24 | Samsung Electronics Company, Ltd. | Reconstruction of three-dimensional video |
-
2016
- 2016-09-30 CN CN201610872000.XA patent/CN106485781A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103236160A (en) * | 2013-04-07 | 2013-08-07 | 水木路拓科技(北京)有限公司 | Road network traffic condition monitoring system based on video image processing technology |
| US20150054913A1 (en) * | 2013-08-21 | 2015-02-26 | Jaunt Inc. | Image stitching |
| US20160088285A1 (en) * | 2014-09-22 | 2016-03-24 | Samsung Electronics Company, Ltd. | Reconstruction of three-dimensional video |
| CN104767975A (en) * | 2015-04-10 | 2015-07-08 | 武汉市测绘研究院 | Method for achieving interactive panoramic video stream map |
| CN104778675A (en) * | 2015-04-28 | 2015-07-15 | 中国矿业大学 | Coal mining fully mechanized excavation face dynamic video image fusion method |
| CN105262949A (en) * | 2015-10-15 | 2016-01-20 | 浙江卓锐科技股份有限公司 | Multifunctional panorama video real-time splicing method |
Non-Patent Citations (1)
| Title |
|---|
| 赵毅力: "多幅图像的自动拼接算法研究", 《计算机工程与应用》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107901424A (en) * | 2017-12-15 | 2018-04-13 | 北京中睿华信信息技术有限公司 | A kind of Image Acquisition modeling |
| CN109978987A (en) * | 2017-12-28 | 2019-07-05 | 周秦娜 | A kind of control method, apparatus and system constructing panorama based on multiple depth cameras |
| CN109147798A (en) * | 2018-07-27 | 2019-01-04 | 北京三快在线科技有限公司 | Audio recognition method, device, electronic equipment and readable storage medium storing program for executing |
| CN109147798B (en) * | 2018-07-27 | 2023-06-09 | 北京三快在线科技有限公司 | Speech recognition method, device, electronic equipment and readable storage medium |
| CN112650422A (en) * | 2020-12-17 | 2021-04-13 | 咪咕文化科技有限公司 | AR interaction method and device of equipment, electronic equipment and storage medium |
| CN115294748A (en) * | 2022-09-08 | 2022-11-04 | 广东中科凯泽信息科技有限公司 | Fixed target disappearance early warning method based on visual data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11410320B2 (en) | Image processing method, apparatus, and storage medium | |
| CN106485781A (en) | Three-dimensional scene construction method based on live video stream and its system | |
| US11330172B2 (en) | Panoramic image generating method and apparatus | |
| Matsuyama et al. | 3D video and its applications | |
| CN106127680B (en) | 720-degree panoramic video fast browsing method | |
| CN101689292B (en) | Banana codec | |
| KR101121034B1 (en) | System and method for obtaining camera parameters from multiple images and computer program products thereof | |
| CN100355272C (en) | Synthesis method of virtual viewpoint in interactive multi-viewpoint video system | |
| CN103198488B (en) | PTZ surveillance camera realtime posture rapid estimation | |
| WO2014117568A1 (en) | Method and device for performing transition between street view images | |
| CN110418146B (en) | Face changing method, storage medium, electronic device and system applied to live scene | |
| CN105069827A (en) | Method for processing video transitions through three-dimensional model | |
| CN106534670B (en) | It is a kind of based on the panoramic video generation method for connecting firmly fish eye lens video camera group | |
| TWI587241B (en) | Method, device and system for generating two - dimensional floor plan | |
| CN107358577B (en) | Rapid splicing method of cubic panoramic image | |
| CN108053373A (en) | One kind is based on deep learning model fisheye image correcting method | |
| CN110689476A (en) | Panoramic image splicing method and device, readable storage medium and electronic equipment | |
| JP2020095717A (en) | Method, system and apparatus for capture of image data for free viewpoint video | |
| CN108629829A (en) | The three-dimensional modeling method and system that one bulb curtain camera is combined with depth camera | |
| CN106023307A (en) | Three-dimensional model rapid reconstruction method and system based on field environment | |
| CN113379901A (en) | Method and system for establishing house live-action three-dimension by utilizing public self-photographing panoramic data | |
| CN116051747A (en) | House three-dimensional model reconstruction method, device and medium based on missing point cloud data | |
| CN103500471A (en) | Method for realizing high-resolution augmented reality system | |
| CN111161399B (en) | Data processing method and assembly for generating three-dimensional model based on two-dimensional image | |
| CN111698412B (en) | Dual-spectrum panoramic image acquisition method and device, unmanned aerial vehicle and client |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170308 |