CN105530505B - 3-D view conversion method and device - Google Patents
3-D view conversion method and device Download PDFInfo
- Publication number
- CN105530505B CN105530505B CN201610027158.7A CN201610027158A CN105530505B CN 105530505 B CN105530505 B CN 105530505B CN 201610027158 A CN201610027158 A CN 201610027158A CN 105530505 B CN105530505 B CN 105530505B
- Authority
- CN
- China
- Prior art keywords
- image block
- image
- converted
- pending
- denoising
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2219/00—Indexing scheme for manipulating 3D models or images for computer graphics
- G06T2219/20—Indexing scheme for editing of 3D models
- G06T2219/2016—Rotation, translation, scaling
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Graphics (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present invention relates to a kind of 3-D view conversion method and device.Wherein, the 3-D view conversion method includes:Pending image to being input into carries out Boundary Recognition, obtains each subject image block in the pending image;Denoising is carried out to each subject image block, each image block to be converted is obtained;Using spindle body Model, the deformation and displacement of each described image block to be converted after denoising are determined, to simulate 3D visual effect of each image block to be converted in the pending image.The present invention is from object deformation, visual effect, only need to identify objects in images, and object is divided, then using spindle body Model as frame of reference, the deformation such as deformation and displacement can be carried out to the object in image according to vision spindle effect, so as to be experienced from the three-dimensional of lifting user.
Description
Technical field
The present invention relates to image processing field, more particularly to a kind of three-dimensional (3Dimensions, 3D) image conversion method and
Device.
Background technology
The broadcasting of increasing video at present plays to show more real visual effect using 3 D stereo.Due to meter
Calculation machine screen is planar, to be shown in computer just like the 3-D view as material object, can be entered video and/image
Conversion of the row from 2 d-to-3 d.Non- 3D films (such as 2D films) are traditionally converted into 3D films, the technology for being used is:Pin
Picture is translated, the algorithm of displacement difference is obtained;Or, using time shaft concept, use time difference image carries out 3D and answers
Position.
But, there is following defect in existing 3D transfer algorithms:Due to that can be translated general image using translation algorithm, cause
Picture is moved integrally, and without distance scape, from the deformation process without true 3D, is moved integrally, and causes the partial content to lack.
Though additionally, it is constant to meet background using time shaft algorithm, 3D effect is manufactured using ohject displacement, vision is easily caused
A series of problems, such as confusion, unintelligible motion, shake, Consumer's Experience is not good.
The content of the invention
Technical problem
In view of this, the technical problem to be solved in the present invention is how to improve the 3D display effects of image.
Solution
In order to solve the above-mentioned technical problem, an a kind of embodiment of the invention, there is provided 3-D view conversion method,
Including:
Pending image to being input into carries out Boundary Recognition, obtains each subject image block in the pending image;
Denoising is carried out to each subject image block, each image block to be converted is obtained;
Using spindle body Model, the deformation and displacement of each described image block to be converted after denoising are determined, to simulate
3D visual effect of each image block to be converted in the pending image.
For the above method, in a kind of possible implementation, the pending image to being input into carries out Boundary Recognition, obtains
Each subject image block in the pending image, including:
Boundary Recognition is carried out to the pending image using thresholding algorithm, each border in the pending image is obtained
Point;
According to the position of each boundary point, each Close edges are determined;
According to each Close edges, each described subject image block in the pending image is determined, wherein the thing
Body image block includes each pixel of a Close edges and its inside.
For the above method, in a kind of possible implementation, the pending image is carried out using thresholding algorithm
Boundary Recognition, obtains each boundary point in the pending image, including:
Grey decision-making to each pixel in the pending image carries out gaussian filtering;
Be defined as the pixel more than in the case of given threshold by the value obtained by a gaussian filtering for pixel
Boundary point.
For the above method, in a kind of possible implementation, denoising is carried out to each subject image block, obtained
To each image block to be converted, including:
Remove background data from the pending image, the background data in the pending image except each institute
State the pixel outside subject image block;
Denoising is carried out to each subject image block, each image block to be converted is obtained.
For the above method, in a kind of possible implementation, using spindle body Model, determine each after denoising
The deformation and displacement of the image block to be converted, including:
For each image block to be converted, the position of the boundary point according to the image block to be converted, it is determined that described wait to turn
Image block is changed in the arc positions residing for the spindle model;
Obtain the corresponding deformation of arc positions and the displacement parameter residing for the spindle model;
According to acquired deformation and displacement parameter, deformation and displacement are carried out to the image block to be converted.
For the above method, in a kind of possible implementation, the pending image be pending video at least
One two field picture.
The present invention also provides a kind of 3-D view conversion equipment, including:
Identification module, for carrying out Boundary Recognition to the pending image being input into, obtains each in the pending image
Subject image block;
Denoising module, is connected with the identification module, for carrying out denoising to each subject image block, obtains each
Image block to be converted;
Three-dimensional modular converter, is connected with the denoising module, for using spindle body Model, determines each after denoising
The deformation and displacement of the image block to be converted, to simulate three-dimensional of each image block to be converted in the pending image
Visual effect.
For said apparatus, in a kind of possible implementation, the identification module includes:
Boundary Recognition unit, for carrying out Boundary Recognition to the pending image using thresholding algorithm, obtains described treating
Each boundary point in treatment image;
Close edges determining unit, is connected with the Boundary Recognition unit, for the position according to each boundary point, really
Fixed each Close edges;
Object determining unit, is connected with the Close edges determining unit, for according to each Close edges, determining institute
Each described subject image block in pending image is stated, wherein the subject image block includes a Close edges and its inside
Each pixel.
For said apparatus, in a kind of possible implementation, the Boundary Recognition unit is additionally operable to wait to locate to described
The grey decision-making of each pixel in reason image carries out gaussian filtering;Value obtained by a gaussian filtering for pixel is more than setting
In the case of determining threshold value, the pixel is defined as boundary point.
For said apparatus, in a kind of possible implementation, the denoising module includes:
Background removal unit, for removing background data from the pending image, the background data is treated for described
Pixel in treatment image in addition to each subject image block;
Denoising unit, for carrying out denoising to each subject image block, obtains each image block to be converted.
For said apparatus, in a kind of possible implementation, the three-dimensional modular converter includes:
Arc positions determining unit, for for each image block to be converted, according to the border of the image block to be converted
The position of point, determines the image block to be converted in the arc positions residing for the spindle model;
Parameter acquiring unit, is connected with the arc positions determining unit, for obtaining residing for the spindle model
The corresponding deformation of arc positions and displacement parameter;
Deformational displacement unit, is connected with the parameter acquiring unit, for according to acquired in the parameter acquiring unit
Deformation and displacement parameter, deformation and displacement are carried out to the image block to be converted.
For said apparatus, in a kind of possible implementation, the pending image be pending video at least
One two field picture.
Beneficial effect
The present invention is from object deformation, visual effect, it is only necessary to identify objects in images, and object is drawn
Point, then using spindle body Model as frame of reference, deformation can be carried out to the object in image according to vision spindle effect
With the deformation such as displacement, so as to from the three-dimensional experience of lifting user.
Additionally, the present invention can be processed the two field picture of the continuous video played, compared with traditional 3D movie conversions,
The present invention is that directly the object on two field picture is processed, and 3D conversions are carried out rather than using displacement difference and time shaft.Cause
This, the present invention can overcome the content caused due to global displacement to lack, it is also possible to overcome due to being caused using time shaft
The problems such as VC, unintelligible motion, shake, with content intact, vision is orderly, flating is small, motion is clear, 3D effects
Really good the advantages of.
According to below with reference to the accompanying drawings to detailed description of illustrative embodiments, further feature of the invention and aspect will become
It is clear.
Brief description of the drawings
Comprising in the description and the part that constitutes specification accompanying drawing together illustrated with specification it is of the invention
Exemplary embodiment, feature and aspect, and for explaining principle of the invention.
Fig. 1 shows the flow chart of 3-D view conversion method according to an embodiment of the invention;
Fig. 2 shows the flow chart of Boundary Recognition process in 3-D view conversion method according to an embodiment of the invention;
Fig. 3 shows the flow chart of denoising process in 3-D view conversion method according to an embodiment of the invention;
Fig. 4 shows the flow chart of three-dimensional transfer process in 3-D view conversion method according to an embodiment of the invention;
Fig. 5 shows the schematic diagram of spindle body Model in 3-D view conversion method according to an embodiment of the invention;
Fig. 6 shows the structured flowchart of 3-D view conversion equipment according to an embodiment of the invention;
Fig. 7 shows the structured flowchart of each module in 3-D view conversion equipment according to an embodiment of the invention.
Specific embodiment
Various exemplary embodiments of the invention, feature and aspect are described in detail below with reference to accompanying drawing.It is identical in accompanying drawing
Reference represent the same or analogous element of function.Although the various aspects of embodiment are shown in the drawings, remove
Non-specifically is pointed out, it is not necessary to accompanying drawing drawn to scale.
Special word " exemplary " means " being used as example, embodiment or illustrative " herein.Here as " exemplary "
Illustrated any embodiment should not necessarily be construed as preferred or advantageous over other embodiments.
In addition, in order to better illustrate the present invention, numerous details are given in specific embodiment below.
It will be appreciated by those skilled in the art that without some details, the present invention can equally be implemented.In some instances, for
Method well known to those skilled in the art, means, element and circuit are not described in detail, in order to highlight purport of the invention.
Embodiment 1
Fig. 1 shows the flow chart of 3-D view conversion method according to an embodiment of the invention.As shown in figure 1, the three-dimensional
Image conversion method can mainly include:
Step 101, the pending image to being input into carry out Boundary Recognition, obtain each object figure in the pending image
As block.
Step 102, denoising is carried out to each subject image block, obtain each image block to be converted.
Step 103, using spindle body Model, determine deformation and the position of each described image block to be converted after denoising
Move, to simulate 3D visual effect of each image block to be converted in the pending image.
Specifically, the image for carrying out 3D conversions will be needed to be referred to as pending image in the embodiment of the present invention.Wherein wait to locate
Reason image can be at least one of static picture, or dynamic video two field picture.The 3-D view conversion side
Method can be performed by CPU or video card in computer etc..
As shown in Fig. 2 in a kind of possible implementation, the Boundary Recognition process of step 101 can specifically include:
Step 201, Boundary Recognition is carried out to pending image using thresholding algorithm, obtain each in the pending image
Boundary point.
Step 202, the position according to each boundary point, determine each Close edges.
Step 203, according to each Close edges, determine each described subject image block in the pending image, its
Described in subject image block include each pixel of a Close edges and its inside.
In a kind of possible implementation, Boundary Recognition is carried out to the pending image using thresholding algorithm, obtained
Each boundary point in the pending image, can specifically include:To the grey decision-making of each pixel in the pending image
Carry out gaussian filtering;In the case that value obtained by a gaussian filtering for pixel is more than given threshold, by the pixel
It is defined as boundary point.
For example, for certain pixel in pending image, by the multiple (such as 1 of the pixel and its surrounding
8 pixels of individual pixel and its surrounding totally 9) pixel grey decision-making (or brightness value, gray value) substitute into Laplce
Gauss operator such as following formula 1 is calculated, and the value being calculated can represent the differentiation of the pixel and surrounding pixel.
Formula 1.
Then, the value that will be calculated is compared with given threshold, if greater than given threshold, represents this pixel
Differentiation degree with surrounding pixel is larger, and this pixel can be judged to the boundary point of object.May in one image
There are multiple objects, the boundary point of usual each object can form Close edges.Wherein Close edges are by pending image
The closed curve that the continuous multiple boundary point in middle position is formed.It is above-mentioned to determine that boundary point is only that one kind is shown using 9 pixels
Example, the present invention does not limit the number of selected specific pixel, according to the difference of the demands such as application scenarios, it would however also be possible to employ its
His number.To circumferential expansion multiple pixel mainly centered on current point, for example, it is also possible to using 25 pixels or not
Number.
For the boundary point that can not form Close edges, can ignore or remove.Each Close edges and its inside it is each
Pixel can represent an object where image block (i.e. subject image block).Certainly it is likely to multiple similar object portions occur
Divide and overlap so that multiple objects are located at the situation inside a Close edges.May be calculated in a pending image many
Individual Close edges, so that it is determined that going out multiple objects image block.
As shown in figure 3, in a kind of possible implementation, the denoising process of step 102 can specifically include:
Step 301, background data is removed from the pending image, during the background data is the pending image
Pixel in addition to each subject image block.
Step 302, denoising is carried out to each subject image block, obtain each image block to be converted.
Specifically, after the subject image block during some pending image is determined, will be except each object
Pixel outside image block is determined as background data, and the background data is removed in subsequent processes, that is to say, that follow-up
No longer the pixel included by background data is calculated during treatment.Additionally, being carried out at denoising for certain subject image block
Reason, for example:Noise detection is carried out using filtering method, while the method denoising judged using the noise on 8 directions, by certain
Noise data removal in subject image block, the image block for finally obtaining is image block to be converted.Denoising is carried out, with side
Boundary is clear, the strong advantage of noise removal capability.
Additionally, in addition to carrying out denoising to each subject image block, it is also possible to pending image is carried out overall
Denoising, such place comprehends more simple and convenient.
As shown in figure 4, in a kind of possible implementation, the three-dimensional transfer process of step 103 can specifically include:
Step 401, for each image block to be converted, the position of the boundary point according to the image block to be converted, it is determined that
The image block to be converted is in the arc positions residing for the spindle model.
The corresponding deformation of arc positions and displacement parameter residing for step 402, the acquisition spindle model.
Step 403, the deformation according to acquired in and displacement parameter, deformation and displacement are carried out to the image block to be converted.
Specifically, spindle body Model 51 as shown in Figure 5 can be previously generated, the spindle body Model has a plurality of arc
Line, each camber line is previously provided with suitable deformation parameter and displacement parameter.Deformation ginseng on every camber line of spindle body Model
Number and displacement parameter are configured according to the vision difference of human eye, and its principle is:Using eyes at horizontal and vertical aspect
Having differences of image objects is projected, and test checking is carried out to projective parameter using position, obtain object deformational displacement ginseng
Number spindle body Model.Therefore, for the object on certain camber line, carried out using the deformation parameter and displacement parameter of the camber line
Deformation and displacement, it becomes possible to make the object that 3D stereoeffects are presented.Furthermore, it is possible to multiple spindle body Models are set, for being adapted to
More scenes.
Additionally, determining that the process of the arc positions residing for object is as follows:It is possible, firstly, to can be with by certain pending picture 53
The center superposition of spindle body Model 51.In such a case, it is possible to the image block each to be converted in pending picture 53 exists
Position in spindle, determines the arc positions residing for each image block to be converted.Preferably, if an image block to be converted 55
The center (such as stain) of (such as the dotted line frame in Fig. 5) and a point weight of a certain bar camber line 57 of spindle body Model 51
Close, it is believed that the image block to be converted 55 is located on this camber line 57.If additionally, the center of image block to be converted does not have
Have and overlapped with the point on any one camber line on the spindle, can be by with the center position of the image block to be converted most
Camber line where near point, is judged to the arc positions residing for the image block to be converted.
After step 103, can be to screen output to the 3-D view after deformation and displacement.
The present invention is from object deformation, visual effect, it is only necessary to identify objects in images, and object is drawn
Point, then using spindle body Model as frame of reference, deformation can be carried out to the object in image according to vision spindle effect
With the deformation such as displacement, so as to from the three-dimensional experience of lifting user.Additionally, the present invention can be to the two field picture of the continuous video played
Processed, compared with traditional 3D movie conversions, the present invention is that directly the object on two field picture is processed, rather than use
Displacement difference and time shaft carry out 3D conversions.Therefore, the present invention can overcome the content caused due to global displacement to lack, and also can
The problems such as enough overcoming VC, unintelligible motion, the shake due to being caused using time shaft, with content intact, vision
In order, the advantages of flating is small, motion is clear, 3D effect is good.
Embodiment 2
Fig. 6 shows the structured flowchart of 3-D view conversion equipment according to an embodiment of the invention.As shown in fig. 6, this three
Dimension image conversion apparatus can mainly include:
Identification module 61, for carrying out Boundary Recognition to the pending image being input into, in obtaining the pending image
Each subject image block;
Denoising module 63, is connected with the identification module 61, for carrying out denoising to each subject image block, obtains
To each image block to be converted;
Three-dimensional modular converter 65, is connected with the denoising module 63, for using spindle body Model, after determining denoising
Each described image block to be converted deformation and displacement, to simulate each image block to be converted in the pending image
3D visual effect.
Each module in the 3-D view conversion equipment of the embodiment of the present invention, is able to carry out the graphics in above-described embodiment
As conversion method.Concrete principle and example may refer to the associated description in above-described embodiment.
As shown in fig. 7, in a kind of possible implementation, the identification module 61 includes:
Boundary Recognition unit 611, for carrying out Boundary Recognition to the pending image using thresholding algorithm, obtains described
Each boundary point in pending image;
Close edges determining unit 613, is connected with the Boundary Recognition unit 611, for according to each boundary point
Position, determines each Close edges;
Object determining unit 615, is connected with the Close edges determining unit 613, for according to each Close edges,
Determine each described subject image block in the pending image, wherein the subject image block include a Close edges and its
Internal each pixel.
In a kind of possible implementation, the Boundary Recognition unit 611 is additionally operable to in the pending image
The grey decision-making of each pixel carries out gaussian filtering;Feelings of the value more than given threshold obtained by a gaussian filtering for pixel
Under condition, the pixel is defined as boundary point.
In a kind of possible implementation, the denoising module 63 includes:
Background removal unit 631, for removing background data from the pending image, the background data is described
Pixel in pending image in addition to each subject image block;
Denoising unit 633, for carrying out denoising to each subject image block, obtains each image to be converted
Block.
In a kind of possible implementation, the three-dimensional modular converter 65 includes:
Arc positions determining unit 651, for for each image block to be converted, according to the side of the image block to be converted
The position of boundary's point, determines the image block to be converted in the arc positions residing for the spindle model;
Parameter acquiring unit 653, is connected with the arc positions determining unit 651, for obtaining the spindle body Model
The corresponding deformation of residing arc positions and displacement parameter;
Deformational displacement unit 655, is connected with the parameter acquiring unit 653, for according to the parameter acquiring unit institute
The deformation of acquisition and displacement parameter, deformation and displacement are carried out to the image block to be converted.
In a kind of possible implementation, the pending image is at least one two field picture of pending video.
3-D view conversion equipment of the invention is from object deformation, visual effect, it is only necessary to identify thing in image
Body, and object is divided, then using spindle body Model as frame of reference, can be according to vision spindle effect to image
In object carry out the deformation such as deformation and displacement, so as to from the three-dimensional experience of lifting user.
Additionally, the present invention can be processed the two field picture of the continuous video played, compared with traditional 3D movie conversions,
The present invention is that directly the object on two field picture is processed, and 3D conversions are carried out rather than using displacement difference and time shaft.Cause
This, the present invention can overcome the content caused due to global displacement to lack, it is also possible to overcome due to being caused using time shaft
The problems such as VC, unintelligible motion, shake, with content intact, vision is orderly, flating is small, motion is clear, 3D effects
Really good the advantages of.
The above, specific embodiment only of the invention, but protection scope of the present invention is not limited thereto, and it is any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (10)
1. a kind of 3-D view conversion method, it is characterised in that including:
Pending image to being input into carries out Boundary Recognition, obtains each subject image block in the pending image;
Denoising is carried out to each subject image block, each image block to be converted is obtained;
Using spindle body Model, the deformation and displacement of each described image block to be converted after denoising are determined, to simulate each institute
State 3D visual effect of the image block to be converted in the pending image;
Wherein, using spindle body Model, the deformation and displacement of each described image block to be converted after denoising are determined, including:
For each image block to be converted, the position of the boundary point according to the image block to be converted determines the figure to be converted
Arc positions as residing for block in the spindle model;
Obtain the corresponding deformation of arc positions and the displacement parameter residing for the spindle model;
According to acquired deformation and displacement parameter, deformation and displacement are carried out to the image block to be converted.
2. method according to claim 1, it is characterised in that the pending image to being input into carries out Boundary Recognition, obtains
Each subject image block in the pending image, including:
Boundary Recognition is carried out to the pending image using thresholding algorithm, each boundary point in the pending image is obtained;
According to the position of each boundary point, each Close edges are determined;
According to each Close edges, each described subject image block in the pending image is determined, wherein the object figure
Include each pixel of a Close edges and its inside as block.
3. method according to claim 2, it is characterised in that row bound is entered to the pending image using thresholding algorithm
Identification, obtains each boundary point in the pending image, including:
Grey decision-making to each pixel in the pending image carries out gaussian filtering;
The pixel is defined as border by the value obtained by a gaussian filtering for pixel more than in the case of given threshold
Point.
4. according to the method in any one of claims 1 to 3, it is characterised in that each subject image block is gone
Make an uproar treatment, obtain each image block to be converted, including:
Remove background data from the pending image, the background data is except each thing in the pending image
Pixel outside body image block;
Denoising is carried out to each subject image block, each image block to be converted is obtained.
5. according to the method in any one of claims 1 to 3, it is characterised in that the pending image is regarded for pending
At least one two field picture of frequency.
6. a kind of 3-D view conversion equipment, it is characterised in that including:
Identification module, for carrying out Boundary Recognition to the pending image being input into, obtains each object in the pending image
Image block;
Denoising module, is connected with the identification module, for carrying out denoising to each subject image block, is respectively waited to turn
Change image block;
Three-dimensional modular converter, is connected with the denoising module, for using spindle body Model, determines each described after denoising
The deformation and displacement of image block to be converted, to simulate 3D vision of each image block to be converted in the pending image
Effect;
Wherein, the three-dimensional modular converter includes:
Arc positions determining unit, for for each image block to be converted, boundary point according to the image block to be converted
Position, determines the image block to be converted in the arc positions residing for the spindle model;
Parameter acquiring unit, is connected with the arc positions determining unit, for obtaining the camber line residing for the spindle model
The corresponding deformation in position and displacement parameter;
Deformational displacement unit, is connected with the parameter acquiring unit, for the deformation according to acquired in the parameter acquiring unit
And displacement parameter, deformation and displacement are carried out to the image block to be converted.
7. device according to claim 6, it is characterised in that the identification module includes:
Boundary Recognition unit, for carrying out Boundary Recognition to the pending image using thresholding algorithm, obtains described pending
Each boundary point in image;
Close edges determining unit, is connected with the Boundary Recognition unit, for the position according to each boundary point, it is determined that respectively
Close edges;
Object determining unit, is connected with the Close edges determining unit, for according to each Close edges, it is determined that described treat
Each described subject image block in treatment image, wherein the subject image block includes each picture of a Close edges and its inside
Vegetarian refreshments.
8. device according to claim 7, it is characterised in that the Boundary Recognition unit is additionally operable to the pending figure
The grey decision-making of each pixel as in carries out gaussian filtering;Value obtained by a gaussian filtering for pixel is more than setting threshold
In the case of value, the pixel is defined as boundary point.
9. the device according to any one of claim 6 to 8, it is characterised in that the denoising module includes:
Background removal unit, for removing background data from the pending image, the background data is described pending
Pixel in image in addition to each subject image block;
Denoising unit, for carrying out denoising to each subject image block, obtains each image block to be converted.
10. the device according to any one of claim 6 to 8, it is characterised in that the pending image is regarded for pending
At least one two field picture of frequency.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610027158.7A CN105530505B (en) | 2016-01-15 | 2016-01-15 | 3-D view conversion method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610027158.7A CN105530505B (en) | 2016-01-15 | 2016-01-15 | 3-D view conversion method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105530505A CN105530505A (en) | 2016-04-27 |
CN105530505B true CN105530505B (en) | 2017-06-23 |
Family
ID=55772446
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610027158.7A Active CN105530505B (en) | 2016-01-15 | 2016-01-15 | 3-D view conversion method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105530505B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107018399A (en) * | 2017-03-28 | 2017-08-04 | 北京犀牛数字互动科技有限公司 | Image data transfer method and device |
CN107102803B (en) * | 2017-04-27 | 2020-06-26 | 北京天耀宏图科技有限公司 | Picture display method and device and computer readable storage medium |
CN108416847B (en) * | 2018-03-09 | 2021-02-19 | 联想(北京)有限公司 | Method and device for displaying operation object |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101075350A (en) * | 2007-06-20 | 2007-11-21 | 中国科学院长春光学精密机械与物理研究所 | Assembly for converting two-dimensional cartoon into three-dimensional cartoon by dynamic outline technology |
US9407904B2 (en) * | 2013-05-01 | 2016-08-02 | Legend3D, Inc. | Method for creating 3D virtual reality from 2D images |
-
2016
- 2016-01-15 CN CN201610027158.7A patent/CN105530505B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN105530505A (en) | 2016-04-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106803067B (en) | Method and device for evaluating quality of face image | |
CN106504271B (en) | Method and apparatus for eye tracking | |
JP4966431B2 (en) | Image processing device | |
WO2017142654A1 (en) | 3-dimensional scene analysis for augmented reality operations | |
US9959600B2 (en) | Motion image compensation method and device, display device | |
JP5310247B2 (en) | Image processing apparatus and method, and program | |
KR20180119696A (en) | Efficient Determination of Optical Flow Between Images | |
Su et al. | Natural scene statistics of color and range | |
US20130136302A1 (en) | Apparatus and method for calculating three dimensional (3d) positions of feature points | |
CN106062824B (en) | edge detecting device and edge detection method | |
CN109859126A (en) | A kind of vedio noise reduction method, apparatus, electronic equipment and storage medium | |
CN115699114A (en) | Image augmentation for analysis | |
CN109711268B (en) | Face image screening method and device | |
US20180189955A1 (en) | Augumented reality (ar) method and system enhanced through efficient edge detection of image objects | |
CN105530505B (en) | 3-D view conversion method and device | |
CN104881843A (en) | Image interpolation method and image interpolation apparatus | |
CN110740309A (en) | image display method, device, electronic equipment and storage medium | |
CN110832542A (en) | Recognition processing device, recognition processing method, and program | |
CN108960012B (en) | Feature point detection method and device and electronic equipment | |
CN113570725A (en) | Three-dimensional surface reconstruction method and device based on clustering, server and storage medium | |
CN108924434A (en) | A kind of three-dimensional high dynamic-range image synthesis method based on exposure transformation | |
JP2012048326A (en) | Image processor and program | |
CN110728692A (en) | Image edge detection method based on Scharr operator improvement | |
CN116342519A (en) | Image processing method based on machine learning | |
Seitner et al. | Trifocal system for high-quality inter-camera mapping and virtual view synthesis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20200508 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Patentee after: Alibaba (China) Co.,Ltd. Address before: 200241, room 2, floor 02, building 555, Dongchuan Road, Minhang District, Shanghai Patentee before: Transmission network technology (Shanghai) Co., Ltd |