CN109859100A - Display methods, electronic equipment and the computer readable storage medium of virtual background - Google Patents
Display methods, electronic equipment and the computer readable storage medium of virtual background Download PDFInfo
- Publication number
- CN109859100A CN109859100A CN201910095071.7A CN201910095071A CN109859100A CN 109859100 A CN109859100 A CN 109859100A CN 201910095071 A CN201910095071 A CN 201910095071A CN 109859100 A CN109859100 A CN 109859100A
- Authority
- CN
- China
- Prior art keywords
- image
- human body
- virtual background
- video
- background
- 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
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of display methods of virtual background.This method comprises: obtaining video image；Human bioequivalence is carried out to the video image, obtains human body image and non-human image；The non-human image is replaced with into preset background image, and the human body image and the background image are overlapped；Image after merging is sent to display end to show.The invention also discloses the display electronic equipments and computer readable storage medium of a kind of virtual background.The present invention can be realized during video, is perfectly combined the virtual background that user selects with real personage by the method for virtual background hypostazation, converts personage's background in video.
Description
Technical field
The present invention relates to field of video image processing more particularly to a kind of display methods of virtual background, electronic equipment and
Computer readable storage medium.
Background technique
With the development of internet technology, more and more communication functions are developed and are applied, wherein video communication function
It can be used widely since the visualization communication of strange land user may be implemented.Currently, video communication function also exists well
Play the role of in terms of medical remote medical consultation with specialists.
It holds a consultation however, doctor is fixed on consulting room.It is normalization that practical doctor job site, which is not fixed: may be in door
It examines and is on diagnostics；It may make the rounds of the wards in department；Community hospital etc. may be descended.In the case of these, doctor can only select to return to consulting room meeting
It examines or is held a consultation on the spot using portable equipment.It is authoritative that the high the latter of the former time cost does not have the consultation of doctors.Therefore the present invention is directed to
Both comprehensive advantage is that the scene matching of consultation of doctor closes the background painting of rule under the premise of authorities allow and authorize
Face and place enrich believable consultation of doctors video mode for patient's presentation is convenient, while also having taken into account protection user in video process
In incoherent other staff personal secrets.
Summary of the invention
The main purpose of the present invention is to provide a kind of display methods of virtual background, electronic equipment and computer-readable deposit
Storage media, it is intended to realize during video, virtual background user select by the method for virtual background hypostazation with show
Real personage is perfectly combined, and converts personage's background in video.
To achieve the above object, the present invention provides a kind of display methods of virtual background, the display side of the virtual background
Method the following steps are included:
Obtain video image；
Human bioequivalence is carried out to the video image, obtains human body image and non-human image；
The non-human image is replaced with into preset background image, and by the human body image and the background image into
Row superposition；
Image after merging is sent to display end to show.
Optionally, described that the non-human image is replaced with into preset background image, and by the human body image and institute
Stating the step of background image is overlapped includes:
Pose calculating is carried out to the human body image of video image, obtains the coordinate of the video image human body image；
According to the coordinate of the human body image, virtual background image is overlapped to human body image.
Optionally, the human body image to video image carries out pose calculating, obtains the video image human body image
Coordinate the step of include:
Obtain human geometry's feature that video image extracts；
The set of eigenvectors formed according to geometrical characteristic parameter；Multiple groups feature vector can confirm the picture human body coordinate；
It is matched according to human body coordinate with the manikin in presetting database, generates characteristics of human body's coordinate pair and answer manikin
Projection matrix；
Position auto―control is calculated according to the projection matrix and preset internal reference matrix, determines that human body coordinate is effective human body
Posture coordinate.
Optionally, the coordinate according to the human body image, the step of virtual background image is overlapped to human body image
Suddenly include:
The preset virtual background image is obtained from presetting database；
The coordinate of the virtual background image and the human body image is subjected to fusion calculation, is superimposed the human body image
Float on virtual background image.
Optionally, the display methods of the virtual background further include:
Obtain the feature vector of human body image in video；
The movement for tracking feature vector in video, judges whether the movement of described eigenvector overruns；
If so, returning to step: obtaining video image.
Optionally, described the step of obtaining the feature vector of human body image in video, includes:
By the projecting video image to multiple lower-dimensional subspaces, image is obtained in the characteristic point of multiple lower-dimensional subspaces；
The characteristic point of the multiple lower-dimensional subspace is combined building, obtains the spy of the human body in the video image
Levy vector.
Optionally, the movement for tracking feature vector in video, judges whether the movement of described eigenvector is more than model
The step of enclosing include:
According to preset kinematic parameter, the motion model of feature vector is calculated；
Whether exceed motion model according to the movement of the eigen vector in video, judge described eigenvector movement whether
It overruns.
Optionally, the preset background image includes two-dimensional virtual background image or three-dimensional background image.
In addition, to achieve the above object, the present invention also provides a kind of display electronic equipment of virtual background, the virtual back
The display electronic equipment of scape includes: memory, processor and is stored on the memory and can run on the processor
Virtual background display program, the display program of the virtual background realizes body as described above when being executed by the processor
The step of display methods of virtual background.
In addition, to achieve the above object, it is described computer-readable the present invention also provides a kind of computer readable storage medium
It is stored with the display program of virtual background on storage medium, is realized when the display program of the virtual background is executed by processor
The step of display methods for the virtual background stated.
The present invention provides display methods, electronic equipment and the computer storage medium of a kind of virtual background.In the method,
Obtain video image；Human bioequivalence is carried out to the video image, obtains human body image and non-human image；It will be described non-human
Image replaces with preset background image, and the human body image and the background image are overlapped；By the figure after merging
It is shown as being sent to display end.By the above-mentioned means, the present invention can be realized during video, pass through human bioequivalence skill
Art identifies human region and background area, and human body image is floated on and is preset on the background image for the scene for closing rule, makes
Background image in video can be changed to the background image of the selection of the user after auditing by administrative department, arrest user not
Beam is that the scene matching of consultation of doctor closes background frame and the place of rule in video place, for patient present it is convenient enrich it is credible
Consultation of doctors video mode, while also taken into account protection user incoherent other staff during video personal secrets.
Detailed description of the invention
Fig. 1 is the apparatus structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to；
Fig. 2 is the flow diagram of the display methods first embodiment of virtual background of the present invention；
Fig. 3 is the flow diagram of the display methods second embodiment of virtual background of the present invention；
Fig. 4 is the flow diagram of the display methods 3rd embodiment of virtual background of the present invention；
Fig. 5 is the flow diagram of the display methods fourth embodiment of virtual background of the present invention；
Fig. 6 is the flow diagram of the 5th embodiment of display methods of virtual background of the present invention；
Fig. 7 is the flow diagram of the display methods sixth embodiment of virtual background of the present invention；
Fig. 8 is the flow diagram of the 7th embodiment of display methods of virtual background of the present invention；
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
As shown in Figure 1, Fig. 1 is the apparatus structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to.
The terminal of that embodiment of the invention can be PC, and being also possible to smart phone, tablet computer, portable computer etc. has number
According to the terminal device of processing function.
As shown in Figure 1, the terminal may include: processor 1001, such as CPU, network interface 1004, user interface
1003, memory 1005, communication bus 1002.Wherein, communication bus 1002 is for realizing the connection communication between these components.
User interface 1003 may include display screen (Display), input unit such as keyboard (Keyboard), optional user interface
1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include that the wired of standard connects
Mouth, wireless interface (such as WI-FI interface).Memory 1005 can be high speed RAM memory, be also possible to stable memory
(non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of aforementioned processor
1001 storage device.
Optionally, terminal can also include camera, RF (Radio Frequency, radio frequency) circuit, sensor, audio
Circuit, Wi-Fi module etc..Wherein, sensor such as optical sensor, motion sensor and other sensors.Specifically, light
Sensor may include ambient light sensor and proximity sensor, wherein ambient light sensor can according to the light and shade of ambient light come
The brightness of display screen is adjusted, proximity sensor can close display screen and/or backlight when mobile terminal is moved in one's ear.As
One kind of motion sensor, gravity accelerometer can detect the size of (generally three axis) acceleration in all directions, quiet
Size and the direction that can detect that gravity when only, the application that can be used to identify mobile terminal posture are (such as horizontal/vertical screen switching, related
Game, magnetometer pose calibrating), Vibration identification correlation function (such as pedometer, tap) etc.；Certainly, mobile terminal can also match
The other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor are set, details are not described herein.
It will be understood by those skilled in the art that the restriction of the not structure paired terminal of terminal structure shown in Fig. 1, can wrap
It includes than illustrating more or fewer components, perhaps combines certain components or different component layouts.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium
Believe module, the display program of Subscriber Interface Module SIM and virtual background.
In terminal shown in Fig. 1, network interface 1004 is mainly used for connecting background server, carries out with background server
Data communication；User interface 1003 is mainly used for connecting client (user terminal), carries out data communication with client；And processor
1001 can be used for calling the display program of the virtual background stored in memory 1005, and execute following operation:
Obtain video image；
Human bioequivalence is carried out to the video image, obtains human body image and non-human image；
The non-human image is replaced with into preset background image, and by the human body image and the background image into
Row superposition；
Image after merging is sent to display end to show.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
Pose calculating is carried out to the human body image of video image, obtains the coordinate of the video image human body image；
According to the coordinate of the human body image, virtual background image is overlapped to human body image.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
Obtain human geometry's feature that video image extracts；
The set of eigenvectors formed according to geometrical characteristic parameter；Multiple groups feature vector can confirm the picture human body coordinate；
It is matched according to human body coordinate with the manikin in presetting database, generates characteristics of human body's coordinate pair and answer manikin
Projection matrix；
Position auto―control is calculated according to the projection matrix and preset internal reference matrix, determines that human body coordinate is effective human body
Posture coordinate.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
The preset virtual background image is obtained from presetting database；
The coordinate of the virtual background image and the human body image is subjected to fusion calculation, is superimposed the human body image
Float on virtual background image.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
Obtain the feature vector of human body image in video；
The movement for tracking feature vector in video, judges whether the movement of described eigenvector overruns；
If so, returning to step: obtaining video image.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
By the projecting video image to multiple lower-dimensional subspaces, image is obtained in the characteristic point of multiple lower-dimensional subspaces；
The characteristic point of the multiple lower-dimensional subspace is combined building, obtains the spy of the human body in the video image
Levy vector.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
According to preset kinematic parameter, the motion model of feature vector is calculated；
Whether exceed motion model according to the movement of the eigen vector in video, judge described eigenvector movement whether
It overruns.
Further, processor 1001 can call the display program of the virtual background stored in memory 1005, also hold
The following operation of row:
The preset background image includes two-dimensional virtual background image or three-dimensional background image.
The specific embodiment of the display equipment of virtual background of the present invention and each embodiment of display methods of following virtual backgrounds
Essentially identical, therefore not to repeat here.
It is the flow diagram of the display methods first embodiment of virtual background of the present invention referring to Fig. 2, Fig. 2, it is described virtual
The display methods of background includes:
Step S100 obtains video image；
The present invention is mainly useful during video, such as during video consultation, naturally it is also possible to be used for it
His video scene.The present embodiment is illustrated for being applied to during video consultation.This implementation process, which mainly passes through, to be had
The equipment of image processing function is completed, which can be equipment PC with image processing function, and server is also possible to
The packaged types terminal device such as smart phone, tablet computer, video recorder, portable computer.Video image can pass through camera
It acquires, can also be obtained by network transmission, camera can be mobile phone camera, computer camera etc..
For example, doctor can mutually establish connection by computer and carry out tele-medicine during video consultation, computer is taken the photograph
As head obtains by video capture the video image of doctor, video image is transmitted to server, server carries out video image
Processing makes the background image of video image replace with the image background of user's selection, is then sent to the other end and is shown, separately
The background image of one end can also be handled similarly, in this way, see between doctor be exactly background image that other side selectes and
It is not true background image, to protect the privacy of both sides.
Step S200 carries out human bioequivalence to the video image, obtains human body image and non-human image；
Human body is identified in the present embodiment, obtains human body image and non-human image, that is, background image.Human bioequivalence,
It is a kind of biological identification technology that the posture feature information based on people carries out identification.Contained with video camera or camera acquisition
The image or video flowing of human body, and automatic detection and tracking human body in the picture, and then the human body detected is identified
A series of the relevant technologies.Human body recognition method can be using the extraction of OPENCV human body contour outline, the human testing based on RGB-D data
Equal recognition methods.By these methods, the identification to human body is identified, obtain human body image and non-human image.
The non-human image is replaced with preset background image by step S300, and by the human body image and described
Background image is overlapped；
Non-human image, that is, background image is replaced with to the background image of user setting, this preset background image includes two
Virtual background image or three-dimensional background image are tieed up, can be that virtual scene file is generated by scene process module,
The picture that can be stored for ground or the background picture provided in webcast website.Image, which merges, can use Logical filtering method, gray scale
The methods of weighted mean method.Replaced background image is merged with human body image, the view after change background can be obtained
Frequency image.
Image after merging is sent to display end and shown by step S400.
Image after merging is passed into display end, can be shown on display end, to show the figure after merging
As being the image after changing background image.The display end can be in the display device on mobile phone, be also possible to the TV of computer
Screen.
The present invention provides display methods, electronic equipment and the computer storage medium of a kind of virtual background.In the method,
Obtain video image；Human bioequivalence is carried out to the video image, obtains human body image and non-human image；It will be described non-human
Image replaces with preset background image, and the human body image and the background image are overlapped；By the figure after merging
It is shown as being sent to display end.By the above-mentioned means, the present invention can be realized during video, pass through human bioequivalence skill
Art identifies human region and background area, and human body image is floated on and is preset on the background image for the scene for closing rule, makes
Background image in video can be changed to the background image of the selection of the user after auditing by administrative department, arrest user not
Beam is that the scene matching of consultation of doctor closes background frame and the place of rule in video place, for patient present it is convenient enrich it is credible
Consultation of doctors video mode, while also taken into account protection user incoherent other staff during video personal secrets.
Referring to Fig. 3, Fig. 3 is the flow diagram of the display methods second embodiment of virtual background of the present invention.
Based on the above embodiment, in the present embodiment, step S300 includes:
Step S310 carries out pose calculating to the human body image of video image, obtains the video image human body image
Coordinate；
In the present embodiment, the projection matrix that characteristics of human body's coordinate pair answers manikin is found according to matching result mapping table；
Position auto―control is calculated according to the projection matrix and preset internal reference matrix, carries out geometry verification and back projection's error point later
Analysis, judges whether it is effective position auto―control.If there is effective position auto―control, it is determined that the seat of the video image
Mark.
Step S320 is overlapped virtual background image to human body image according to the coordinate of the human body image.
After determining the coordinate of the video image, the coordinate of background image in defining database, according to two seats
Mark is overlaped, i.e., background image rendering is added in video image by calculating to merge on two coordinates.
Referring to Fig. 4, Fig. 4 is the flow diagram of the display methods 3rd embodiment of virtual background of the present invention.
Based on the above embodiment, in the present embodiment, step S310 includes:
Step S311 obtains human geometry's feature that video image extracts；
In the present embodiment, video image human body coordinate is obtained, which can be special by carrying out human geometry to video image
Sign is extracted, the multiple groups set of eigenvectors confirmation of formation.
Step S312, the set of eigenvectors formed according to geometrical characteristic parameter；Multiple groups feature vector can confirm picture people
Body coordinate；
In the present embodiment, the projection matrix of human body (2D coordinate) corresponding manikin (3D coordinate) in video image is obtained,
2 dimension images in video image and 3 dimension manikins in database are established into projection matrix, obtain video image to database
The corresponding relationship of middle manikin.
Step S313 is matched according to human body coordinate with the manikin in presetting database, generates characteristics of human body's coordinate pair
Answer the projection matrix of manikin；
Restore position auto―control according to the internal reference matrix of projection matrix and camera internal, i.e., according to the projection matrix and default
Internal reference matrix generate position auto―control.
Step S314 calculates position auto―control according to the projection matrix and preset internal reference matrix, determines human body coordinate
For effective human body attitude coordinate.
After obtaining position auto―control, geometry verification and back projection's error analysis are carried out, judges whether it is effective pose
Matrix.If there is effective position auto―control, then the coordinate of the video image is obtained.Geometry verification and back projection's error point
Analysis is common technology means, is not described in detail herein.
Referring to Fig. 5, Fig. 5 is the flow diagram of the display methods fourth embodiment of virtual background of the present invention.
Based on the above embodiment, in the present embodiment, step S320 includes:
Step S321 obtains the preset virtual background image from presetting database；
From preset tranining database, the preset virtual background image that user selectes is obtained, by virtual background image
Define a coordinate.Make the reference body that virtual background image one is static, convenient for being superimposed personage's head in virtual background image
Picture.
The coordinate of the virtual background image and the human body image is carried out fusion calculation, makes the people by step S322
Body image superposition floats on virtual background image.
After the coordinate for obtaining preset virtual background image and video image, by dummy object (virtual background image) with
The alignment of real scene coordinate system, and carry out the fusion calculation of virtual scene；Human body image superposition is set to float on virtual background image
On, the information after system globe area can display in real time in the display.In this way, by virtual background image, it can be by human body image
Superposition floats on virtual background image.
Referring to Fig. 6, Fig. 6 is the flow diagram of the 5th embodiment of display methods of virtual background of the present invention.
Based on the above embodiment, the present embodiment step further include:
Step S500 obtains the feature vector of human body image in video；
Video image is obtained, image information is converted into eigen vector information.Mainly positioned using fusion human body three-dimensional
With augmented reality by carrying out multiple space dimensionality reductions and feature extraction to video image, extract as more as possible in picture human body
Characteristic value, construction feature vector.Feature vector correspondence includes the characteristic information of human body in picture.Different people in picture
Body, corresponding feature vector is different, and each feature vector is equivalent to capture or describes a kind of variation or spy between human body
Property, each human body may be expressed as the linear combination of these characteristic values.Thus it can go out human body area using eigenvector recognition
Domain.
Step S600 tracks the movement of feature vector in video, judges whether the movement of described eigenvector overruns；
If so, S100 is returned to step: obtain video image.
In the present embodiment, by detecting position auto―control, characteristic point is extracted, penalty coefficient is solved, judges the characteristic point of tracking
Position auto―control precision whether is influenced, the case where influence, second point is returned to and re-recognizes matching rendering.Furtherly, it exactly images
Head continues the picture of acquisition, and background service carries out tracking calculating to the picture, when the points traced into have influenced calculating pose
When the precision of matrix, then re-starts recognition detection and render again, it is reciprocal with this.I.e. camera continues capturing visual, and backstage takes
The key point of picture, and the motion model created in real time based on interframe continuity are extracted in business, export pose on the basis of the two herein
Matrix target value can assert image motion, then need to re-start recognition detection when the key point traced into exceeds target value
It renders again, reciprocal with this, to reach that background is different according to physicians location posture and Dynamically Announce, doctor is as really staying
In virtual background.
Referring to Fig. 7, Fig. 7 is the flow diagram of the display methods sixth embodiment of virtual background of the present invention.
Based on the above embodiment, in the present embodiment, step S500 includes:
The projecting video image to multiple lower-dimensional subspaces is obtained image in multiple lower-dimensional subspaces by step S510
Characteristic point；
Basic ideas in this implementation are, two-dimensional image elder generation gray processing, to be converted into the image in a channel, later again
It is end to end to be converted into a column vector, it is assumed that image size is 20*20, then this vector is exactly 400 dimensions, theoretically
It says and is organized into a vector, so that it may apply any machine learning algorithm, but the too high algorithm complexity of dimension also can be therewith
It increases, so needing using PCA algorithm dimensionality reduction, then can be carried out calculating using simple sequence or KNN.Specifically,
Original image can be converted to the original image of N-dimensional vector, project image onto the feature space that Spatial Dimension is less than N, will be empty
Between dimension less than the feature space of N be defined as lower-dimensional subspace, the position and projection line by its subpoint in subspace
Length determine image in the characteristic point in multiple subspaces.After image is transformed to another space, the other figure of same class
As coming together, different classes of image convergent force is distant, and different classes of image is in distribution in original pixel space
It is difficult to transform to another space, so that it may well separate them with simple line or face cutting.It is more by projecting
A lower-dimensional subspace, so that it may using original image the characteristic point of different subspace difference, so that characteristics of image is subject to
Difference classification, can be obtained by image in the characteristic point of multiple subspaces in this way.The present embodiment can be using PCA (principal component point
Analysis) spatial transform method, the main component of human body distribution is obtained using PCA, specific implementation is to human figures all in training set
The covariance matrix of picture carries out eigen value decomposition, obtains corresponding eigenvector, i.e. characteristic point.
The characteristic point of the multiple lower-dimensional subspace is combined building, obtained in the video image by step S520
Human body feature vector.
By according to the sequence arrayed feature vector from big to small of characteristic value, the feature vector corresponding to maximum eigenvalue is anti-
The maximum difference between training image is answered, and the feature vector that corresponding characteristic value is smaller, the difference between the image of reaction is smaller.
The corresponding feature vector of all nonzero eigenvalues, composition characteristic space, that is, so-called " eigenface " space, arrayed feature
Vector: according to nonzero eigenvalue, sequence from big to small arranges corresponding feature vector.Composed eigenvectors matrix
The each of as feature space U, U is classified as a feature vector.Width human body image each in this way can project to by
In subspace.Therefore each width human body image corresponds to a point in subspace, equally, each pair of point Ying Yu in subspace
Piece image.Training image projects to obtain eigenface subspace, by such a dimensionality reduction feature being made of " eigenface "
Space, the human body image after any width centralization can be projected to eigenface subspace by following formula and obtain one
Group coordinate system includes the eigen vector of human body information.
Referring to Fig. 8, Fig. 8 is the flow diagram of the 7th embodiment of display methods of virtual background of the present invention.
Based on the above embodiment, in the present embodiment, step S600 includes:
Step S610 calculates the motion model of feature vector according to preset kinematic parameter；
The present embodiment, tracking phase carries out tracking and calculates new position auto―control, when the points traced into have influenced calculating
When the precision of position auto―control, identification and detection are re-started.The track side of more key frames (Multi Keyframe) can be used
Formula estimates the globe motion parameter of pose by edge contour feature extraction, in conjunction with the motion model of pose, solves
The penalty coefficient of exercise parameter is realized and is solved to the random scanning and globe motion parameter of pose, is carried out to pose adaptive
Harris Corner Detection, the pose target value after finally output detection, realization are quickly tracked and are identified to the key frame of pose.Position
Appearance target value is achieved with the motion model of feature vector.According to the possible fortune of moving description feature vector of feature vector
Dynamic rule.
Whether step S620 exceeds motion model according to the movement of the eigen vector in video, judges described eigenvector
Movement whether overrun.
Firstly, carrying out image noise reduction and Corner Detection processing to the original image of input, pass through the overall situation to exercise parameter
Estimation and error compensation, realization quickly track and identify the key frame of moving image.To the key frame tracking of moving image
Mainly there are the frame tracking based on the fusion of white balance deviation, the key frame tracking based on LWT wavelet division, based on movement
The key frame tracking for the Estimation of Parameters that decays.By continuing to obtain video image, by feature vector in video image
Movement is tracked, and whether is exceeded motion model according to the movement of the eigen vector in video, is judged the fortune of described eigenvector
It is dynamic whether to overrun.For example, the key point of picture is basic position auto―control, by motion model, object pose matrix is calculated,
Just as static (0,0) coordinate, motion model be it is per second walk 1m, then being equivalent to away one when target value is your (0,1) coordinate
Step.
The concrete principle of the present embodiment is exactly: moving image is in collection process, usually by the interference of noise, need into
The noise reduction process of row image improves the tracking accuracy to the key frame of target image, on the basis of image filtering noise reduction process
On, gray proces are carried out to moving image, the globe motion parameter of moving image is estimated by edge contour feature extraction
Meter solves the compensation system of exercise parameter in conjunction with the motion model of image, realizes and joins to the random scanning and global motion of image
Number solves, and on this basis, carries out adaptive H arris Corner Detection to moving image, the moving target after finally output detection
Value realizes that the key frame express delivery to moving image is tracked and identified.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium.
The display program of virtual background, the display of the virtual background are stored on computer readable storage medium of the present invention
The step of display methods of virtual background as described above is realized when program is executed by processor.
Wherein, the display program of the virtual background run on the processor is performed realized method and can refer to
The each embodiment of the display methods of virtual background of the present invention, details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of display methods of virtual background, which is characterized in that the display methods of the virtual background the following steps are included:
Obtain video image；
Human bioequivalence is carried out to the video image, obtains human body image and non-human image；
The non-human image is replaced with into preset background image, and the human body image and the background image are folded
Add；
Image after merging is sent to display end to show.
2. the display methods of virtual background as described in claim 1, which is characterized in that described to replace the non-human image
For preset background image, and the step of human body image and the background image are overlapped includes:
Pose calculating is carried out to the human body image of video image, obtains the coordinate of the video image human body image；
According to the coordinate of the human body image, virtual background image is overlapped to human body image.
3. the display methods of virtual background as claimed in claim 2, which is characterized in that the human body attitude to video image
Identification, the step of obtaining the coordinate of the video image human body attitude include:
Obtain human geometry's feature that video image extracts；
The set of eigenvectors formed according to geometrical characteristic parameter；Multiple groups feature vector can confirm the picture human body coordinate；
It is matched according to human body coordinate with the manikin in presetting database, generates the throwing that characteristics of human body's coordinate pair answers manikin
Shadow matrix；
Position auto―control is calculated according to the projection matrix and preset internal reference matrix, determines that human body coordinate is effective human body attitude
Coordinate.
4. the display methods of virtual background as claimed in claim 2, which is characterized in that the seat according to the human body image
Mark, the step of being overlapped virtual background image to human body image include:
The preset virtual background image is obtained from presetting database；
The coordinate of the virtual background image and the human body image is subjected to fusion calculation, floats on the human body image superposition
On virtual background image.
5. the display methods of virtual background as described in claim 1, which is characterized in that the display methods of the virtual background is also
Include:
Obtain the feature vector of human body image in video；
The movement for tracking feature vector in video, judges whether the movement of described eigenvector overruns；
If so, returning to step: obtaining video image.
6. the display methods of virtual background as claimed in claim 5, which is characterized in that described to obtain human body image in video
The step of feature vector includes:
By the projecting video image to multiple lower-dimensional subspaces, image is obtained in the characteristic point of multiple lower-dimensional subspaces；
The characteristic point of the multiple lower-dimensional subspace is combined building, obtain the feature of the human body in the video image to
Amount.
7. the display methods of virtual background as claimed in claim 5, which is characterized in that feature vector in the tracking video
Movement, judges that the step of whether movement of described eigenvector overruns includes:
According to preset kinematic parameter, the motion model of feature vector is calculated；
Whether exceed motion model according to the movement of the eigen vector in video, judge described eigenvector movement whether be more than
Range.
8. the display methods of virtual background as described in claim 1, which is characterized in that the preset background image includes two
Tie up virtual background image or three-dimensional background image.
9. a kind of display electronic equipment of virtual background, which is characterized in that the display electronic equipment of the virtual background includes: to deposit
Reservoir, processor and the display program for being stored in the virtual background that can be run on the memory and on the processor, institute
It states and realizes the virtual background as described in any one of claims 1 to 8 when the display program of virtual background is executed by the processor
Display methods the step of.
10. a kind of computer readable storage medium, which is characterized in that be stored with virtual back on the computer readable storage medium
The display program of scape is realized when the display program of the virtual background is executed by processor such as any one of claims 1 to 8 institute
The step of stating the display methods of virtual background.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910095071.7A CN109859100A (en) | 2019-01-30 | 2019-01-30 | Display methods, electronic equipment and the computer readable storage medium of virtual background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910095071.7A CN109859100A (en) | 2019-01-30 | 2019-01-30 | Display methods, electronic equipment and the computer readable storage medium of virtual background |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109859100A true CN109859100A (en) | 2019-06-07 |
Family
ID=66897099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910095071.7A Pending CN109859100A (en) | 2019-01-30 | 2019-01-30 | Display methods, electronic equipment and the computer readable storage medium of virtual background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859100A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111901552A (en) * | 2020-06-29 | 2020-11-06 | 维沃移动通信有限公司 | Multimedia data transmission method and device and electronic equipment |
| CN111901672A (en) * | 2020-06-12 | 2020-11-06 | 深圳市京华信息技术有限公司 | Artificial intelligence image processing method |
| CN113012042A (en) * | 2019-12-20 | 2021-06-22 | 海信集团有限公司 | Display device, virtual photo generation method, and storage medium |
| CN114584704A (en) * | 2022-02-08 | 2022-06-03 | 维沃移动通信有限公司 | Shooting method and device and electronic equipment |
| WO2023015868A1 (en) * | 2021-08-13 | 2023-02-16 | 华为技术有限公司 | Image background generation method and aparatus, and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120044260A1 (en) * | 2010-08-18 | 2012-02-23 | Jun Hirai | Image Processing Device, Method, and Program |
| CN107231529A (en) * | 2017-06-30 | 2017-10-03 | 努比亚技术有限公司 | Image processing method, mobile terminal and storage medium |
| CN107592491A (en) * | 2017-09-11 | 2018-01-16 | 广东欧珀移动通信有限公司 | Video communication background display methods and device |
| CN107613239A (en) * | 2017-09-11 | 2018-01-19 | 广东欧珀移动通信有限公司 | Video communication background display methods and device |
-
2019
- 2019-01-30 CN CN201910095071.7A patent/CN109859100A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120044260A1 (en) * | 2010-08-18 | 2012-02-23 | Jun Hirai | Image Processing Device, Method, and Program |
| CN107231529A (en) * | 2017-06-30 | 2017-10-03 | 努比亚技术有限公司 | Image processing method, mobile terminal and storage medium |
| CN107592491A (en) * | 2017-09-11 | 2018-01-16 | 广东欧珀移动通信有限公司 | Video communication background display methods and device |
| CN107613239A (en) * | 2017-09-11 | 2018-01-19 | 广东欧珀移动通信有限公司 | Video communication background display methods and device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113012042A (en) * | 2019-12-20 | 2021-06-22 | 海信集团有限公司 | Display device, virtual photo generation method, and storage medium |
| CN113012042B (en) * | 2019-12-20 | 2023-01-20 | 海信集团有限公司 | Display device, virtual photo generation method, and storage medium |
| CN111901672A (en) * | 2020-06-12 | 2020-11-06 | 深圳市京华信息技术有限公司 | Artificial intelligence image processing method |
| CN111901552A (en) * | 2020-06-29 | 2020-11-06 | 维沃移动通信有限公司 | Multimedia data transmission method and device and electronic equipment |
| CN111901552B (en) * | 2020-06-29 | 2022-10-25 | 维沃移动通信有限公司 | Multimedia data transmission method and device and electronic equipment |
| WO2023015868A1 (en) * | 2021-08-13 | 2023-02-16 | 华为技术有限公司 | Image background generation method and aparatus, and computer-readable storage medium |
| CN114584704A (en) * | 2022-02-08 | 2022-06-03 | 维沃移动通信有限公司 | Shooting method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kim et al. | Revisiting trends in augmented reality research: A review of the 2nd decade of ISMAR (2008–2017) | |
| US11960651B2 (en) | Gesture-based shared AR session creation | |
| CN109859100A (en) | Display methods, electronic equipment and the computer readable storage medium of virtual background | |
| KR102069964B1 (en) | Virtual reality-based apparatus and method to generate a three dimensional(3d) human face model using image and depth data | |
| EP3798801A1 (en) | Image processing method and apparatus, storage medium, and computer device | |
| KR101979669B1 (en) | Method for correcting user’s gaze direction in image, machine-readable storage medium and communication terminal | |
| CN102270275B (en) | The method of selecting object and multimedia terminal in virtual environment | |
| CN114981844A (en) | 3D body model generation | |
| US11417052B2 (en) | Generating ground truth datasets for virtual reality experiences | |
| US20150042640A1 (en) | Floating 3d image in midair | |
| US20120200667A1 (en) | Systems and methods to facilitate interactions with virtual content | |
| CN105210093A (en) | Devices, systems and methods of capturing and displaying appearances | |
| CN109859857A (en) | Mask method, device and the computer readable storage medium of identity information | |
| CN110458924B (en) | Three-dimensional face model establishing method and device and electronic equipment | |
| CN112954292A (en) | Digital museum navigation system and method based on augmented reality | |
| US20210304429A1 (en) | Depth estimation using biometric data | |
| US20210201502A1 (en) | Method and system for motion prediction | |
| CN111308707A (en) | Picture display adjusting method and device, storage medium and augmented reality display equipment | |
| KR20230027237A (en) | Reconstruction of 3D object models from 2D images | |
| CN105814604B (en) | For providing location information or mobile message with the method and system of at least one function for controlling vehicle | |
| KR101864717B1 (en) | The apparatus and method for forming a augmented reality contents with object shape | |
| Pinhanez et al. | Intelligent studios: Using computer vision to control TV cameras | |
| CN114358112A (en) | Video fusion method, computer program product, client and storage medium | |
| CN107066095B (en) | Information processing method and electronic equipment | |
| US11282282B2 (en) | Virtual and physical reality integration |
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 |