CN109636926B - 3D global free deformation method and device - Google Patents
3D global free deformation method and device Download PDFInfo
- Publication number
- CN109636926B CN109636926B CN201811403638.4A CN201811403638A CN109636926B CN 109636926 B CN109636926 B CN 109636926B CN 201811403638 A CN201811403638 A CN 201811403638A CN 109636926 B CN109636926 B CN 109636926B
- Authority
- CN
- China
- Prior art keywords
- image
- target
- point
- stored
- global free
- 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
Images
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
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
Landscapes
- Engineering & Computer Science (AREA)
- Architecture (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 Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a 3D global free deformation algorithm and a device, wherein the 3D global free deformation algorithm comprises the following steps: acquiring a 3D image; matching the 3D image with a pre-stored image library, wherein each pre-stored image in the pre-stored image library is provided with a tensor model for adjusting the shape of the image point of the pre-stored image; generating a 3D image with a tensor model arranged on an image point by utilizing the pre-stored tensor model on the image through an artificial intelligence deep learning algorithm; and deforming the image points on the 3D image through a tensor model. The 3D global free deformation algorithm and the device can obtain more standard digital point cloud with adjustable shape, so that the obtained 3D image is easier to manage and control, and resources consumed by operation can be reduced.
Description
Technical Field
The invention relates to a 3D global free deformation method and device.
Background
The 3D camera, which is a camera manufactured by using a 3D lens, generally has two or more imaging lenses, and the distance between the two imaging lenses is close to the distance between human eyes, so that different images of the same scene seen by similar human eyes can be captured. The holographic 3D has a disc 5 above the lens.
The first 3D camera to date the 3D revolution has all been around the hollywood heavy-pound large and major sporting events. With the advent of 3D cameras, this technology is one step closer to home users. After the camera is introduced, each memorable moment of the life, such as the first step taken by a child, a university graduation celebration and the like, can be captured by using a 3D lens in the future.
A 3D camera typically has more than two lenses. The 3D camera functions like a human brain, and can fuse two lens images together to form a 3D image. These images can be played on a 3D television, and can be viewed by viewers wearing so-called actively shuttered glasses, or directly viewed by naked-eye 3D display devices. The 3D shutter glasses can rapidly stagger the opening and closing of the lenses of the left and right glasses at a rate of 60 times per second. This means that each eye sees a slightly different picture of the same scene, so the brain can thus think that it is enjoying a single picture in 3D.
The existing 3D camera has the defect that the images acquired by the camera cannot be controlled.
Disclosure of Invention
The invention aims to overcome the defect that images acquired by a 3D camera are not easy to process and control in the prior art, and provides a 3D global free deformation method and a device which can acquire more standard digital point clouds and enable the acquired 3D images to be easier to manage and control.
The invention solves the technical problems through the following technical scheme:
A3D global free deformation method, characterized in that the 3D global free deformation method comprises:
acquiring a 3D image;
matching the 3D image with a pre-stored image library, wherein each pre-stored image in the pre-stored image library is provided with a tensor model for adjusting the shape of the image point of the pre-stored image;
generating a 3D image with a tensor model arranged on an image point by utilizing the pre-stored tensor model on the image through an artificial intelligence deep learning algorithm;
and the image points on the 3D image provided with the tensor model on the image points are deformed through the tensor model.
The 3D image is a face image.
Preferably, the tensor model is a function expression which is set on a prestored image and represents the relationship between image points, and the 3D global free deformation method includes:
and setting a function formula between image points on the 3D image by utilizing the pre-stored function formula on the image through an artificial intelligence deep learning algorithm.
Preferably, each pre-stored image in the pre-stored image library is divided into a plurality of regions, and each region is provided with a function expression representing the relationship between image points in the same region, and the 3D global free deformation method includes:
dividing regions on the 3D image by using the region positions on the pre-stored image through an artificial intelligence deep learning algorithm;
and for a target area on the 3D image, setting a function between image points in the target area on the 3D image by utilizing the pre-stored function on the image through an artificial intelligence deep learning algorithm.
Preferably, the 3D global free deformation method includes:
for a target pre-stored image in a pre-stored image library, acquiring a function formula between adjacent image points in the target pre-stored image, wherein the function formula is a polynomial function;
and obtaining a plurality of parting lines passing through adjacent image points through artificial intelligence deep learning, calculating the sum of times of the highest-order item of a polynomial function between all the adjacent image points passing through the parting lines, and dividing the region of the target pre-stored image by the parting lines with the sum of times lower than a preset value.
Preferably, the 3D global free deformation method includes:
matching the 3D image with a target image in the pre-stored image library;
adjusting the space shape of the target image according to the space shape of the 3D image through an artificial intelligence deep learning algorithm;
and taking the target image with the adjusted space shape as a 3D image with a tensor model on the image point.
Preferably, the 3D global free deformation method includes:
overlapping the 3D image and the target image to obtain the distance from an image point on the target image to the 3D image;
acquiring the image point with the maximum distance as a control point, and moving the control point to the direction of the 3D image by the target length;
and moving the surrounding control points around the control point to the direction of the 3D image by using a tensor model of the target image to calculate the length, wherein the size of the calculated length of each surrounding control point is in inverse proportion to the distance from the surrounding control point to the control point, and the calculated length is less than the target length.
Preferably, the deforming the image point on the 3D image with the tensor model on the image point by the tensor model includes:
acquiring an adjusting instruction for adjusting a target image point on the 3D image;
moving the target image point to a target direction by an instruction length according to an adjusting instruction;
and moving the peripheral image points around the target image point to the target direction by utilizing the tensor model of the 3D image to calculate the length, wherein the size of the calculated length of each peripheral image point is in inverse proportion to the distance from the peripheral image point to the target image point, and the calculated length is less than the target length.
The invention also provides a 3D global free deformation device, which is characterized in that the 3D global free deformation device is used for realizing the 3D global free deformation method.
On the basis of the common knowledge in the field, the above preferred conditions can be combined randomly to obtain the preferred embodiments of the invention.
The positive progress effects of the invention are as follows:
the 3D global free deformation method and the device can obtain more standard digital point cloud with adjustable shape, so that the obtained 3D image is easier to manage and control, and resources consumed by operation can be reduced.
Drawings
Fig. 1 is a flowchart of a 3D global free deformation method according to embodiment 1 of the present invention.
Fig. 2 is a flowchart of a 3D global free deformation method according to embodiment 2 of the present invention.
Detailed Description
The invention is further illustrated by the following examples, which are not intended to limit the invention thereto.
Example 1
The embodiment provides a 3D global free deformation apparatus, where the 3D global free deformation apparatus includes an obtaining module, a matching module, a generating module, and a processing module.
The acquisition module is used for acquiring a 3D image, and the 3D image is a human face image in the embodiment.
The matching module is used for matching the 3D image with a pre-stored image library, and each pre-stored image in the pre-stored image library is provided with a tensor model for adjusting the shape of the image point of the pre-stored image;
the generation module is used for generating a 3D image with a tensor model arranged on an image point by utilizing the pre-stored image tensor model through an artificial intelligence deep learning algorithm;
the processing module is used for enabling the image points on the 3D image with the tensor model on the image points to deform through the tensor model.
Machine learning is achieved through an algorithm, so that a machine can learn rules from a large amount of data input from the outside, and recognition and judgment are carried out. The method and the device have the advantages that the images in the standard images (pre-stored image libraries) are learned, the rules of the images in the pre-stored image libraries are obtained, the fluctuation rules of the face model can be obtained, if the curve of the nose and the model relation existing in the tip of the nose exist, a tensor model can be established through the relation between digital points, and the tensor is a multi-linear function which can be used for expressing the linear relation among some vectors, scalars and other tensors.
Specifically, the tensor model is a function formula which is set on a prestored image and represents the relation between image points, and the generation module is used for setting the function formula between the image points on the 3D image by utilizing the function formula on the prestored image through an artificial intelligence deep learning algorithm.
Furthermore, each pre-stored image in the pre-stored image library is divided into a plurality of areas, a function expression for representing the relation between image points in the same area is arranged in each area, and the 3D global free deformation device comprises a division module.
The dividing module is used for dividing the region on the 3D image by utilizing the region position on the pre-stored image through an artificial intelligence deep learning algorithm;
for a target area on the 3D image, the generating module is used for setting a function between image points in the target area on the 3D image by utilizing the function on the prestored image through an artificial intelligence deep learning algorithm.
Because the relation between each image point is very complicated, if the image points which are involved in the movement of one image point are calculated from the whole, the calculation amount is very huge, so that the image points with obvious linkage relation are divided into the same area, the relation with the image points outside the area is cut off, and the calculation amount can be reduced.
The 3D global free deformation apparatus of the present embodiment is also used for dividing regions. The 3D global free deformation device also comprises a calculation module.
For a target pre-stored image in a pre-stored image library, the calculation module is used for acquiring a function formula between adjacent image points in the target pre-stored image, and the function formula is a polynomial function;
the calculation module is also used for acquiring a plurality of segmentation lines passing through adjacent image points through artificial intelligence deep learning, calculating the sum of times of the highest-order item of the polynomial function between all the adjacent image points through which the segmentation lines pass, and dividing the region of the target pre-stored image by the segmentation lines of which the sum of times is lower than a preset value.
The 3D global free deformation device is also used for controlling the deformation of the 3D image added with the tensor model, and the 3D global free deformation device further comprises a receiving module and an adjusting module.
The receiving module is used for acquiring an adjusting instruction for adjusting a target image point on the 3D image;
the adjusting module is used for moving the target image point to a target direction by an instruction length according to an adjusting instruction;
the adjusting module is further configured to move the peripheral image points around the target image point to the target direction by a calculated length using the tensor model of the 3D image, the calculated length of each peripheral image point is inversely proportional to a distance from the peripheral image point to the target image point, and the calculated length is smaller than the target length.
Referring to fig. 1, with the above 3D global free deformation apparatus, this embodiment further provides a 3D global free deformation method, including:
each pre-stored image in the pre-stored image library is divided into a plurality of areas, and a function expression for representing the relation between image points in the same area is arranged in each area.
102, dividing regions on the 3D image by using the region positions on the pre-stored image through an artificial intelligence deep learning algorithm;
specifically, step 102 includes: for a target pre-stored image in a pre-stored image library, acquiring a function formula between adjacent image points in the target pre-stored image, wherein the function formula is a polynomial function;
then, step 102 further includes obtaining a plurality of dividing lines passing through between adjacent image points through artificial intelligence deep learning, calculating the sum of times of the highest-order term of the polynomial function between all adjacent image points passing through the dividing lines, and dividing the region of the target pre-stored image by the dividing lines with the sum of times lower than a preset value.
103, setting a function formula between image points in a target area on the 3D image by utilizing the function formula on the pre-stored image through an artificial intelligence deep learning algorithm.
Step 103 is utilized to realize the generation of the 3D image with the tensor model on the image points by utilizing the pre-stored tensor model on the image through the artificial intelligence deep learning algorithm.
104, acquiring an adjusting instruction for adjusting a target image point on the 3D image;
105, moving the target image point to a target direction by an instruction length according to an adjusting instruction;
and 106, moving the peripheral image points around the target image point to the target direction by using the tensor model of the 3D image to calculate the length, wherein the calculated length of each peripheral image point is inversely proportional to the distance from the peripheral image point to the target image point, and the calculated length is less than the target length.
Through the steps 104 to 106, the deformation of the image point on the 3D image with the tensor model on the image point is realized through the tensor model.
The 3D global free deformation method and the device can acquire more standard digital point clouds with adjustable shapes, so that the acquired 3D images are easier to manage and control, and resources consumed by calculation can be reduced.
Example 2
This embodiment is substantially the same as embodiment 1 except that:
the 3D global free deformation apparatus of this embodiment includes a matching module, an adjusting module, and a setting module.
The matching module is used for matching the 3D image with a target image in the pre-stored image library;
the adjusting module is used for adjusting the space shape of the target image according to the space shape of the 3D image through an artificial intelligence deep learning algorithm;
the setting module is used for taking the target image with the adjusted space shape as the 3D image with the tensor model on the image points.
The adjusting module is specifically configured to:
overlapping the 3D image and the target image to obtain the distance from an image point on the target image to the 3D image;
acquiring the image point with the maximum distance as a control point, and moving the control point to the direction of the 3D image by the target length;
and moving the surrounding control points around the control point to the direction of the 3D image by using a tensor model of the target image to calculate the length, wherein the size of the calculated length of each surrounding control point is in inverse proportion to the distance from the surrounding control point to the control point, and the calculated length is less than the target length.
Correspondingly, the 3D global free deformation method of the present embodiment includes:
and each pre-stored image in the pre-stored image library is provided with a tensor model for adjusting the shape of the image point of the pre-stored image.
And 202, adjusting the space shape of the target image according to the space shape of the 3D image through an artificial intelligence deep learning algorithm.
Step 202 specifically includes:
overlapping the 3D image and the target image to obtain the distance from an image point on the target image to the 3D image;
acquiring the image point with the maximum distance as a control point, and moving the control point to the direction of the 3D image by the target length;
and moving the surrounding control points around the control point to the direction of the 3D image by using a tensor model of the target image to calculate the length, wherein the size of the calculated length of each surrounding control point is in inverse proportion to the distance from the surrounding control point to the control point, and the calculated length is less than the target length.
And 203, taking the target image with the adjusted space shape as a 3D image with a tensor model on the image point.
And 204, deforming the image points on the 3D image with the tensor models on the image points through the tensor models.
While specific embodiments of the invention have been described above, it will be appreciated by those skilled in the art that these are by way of example only, and that the scope of the invention is defined by the appended claims. Various changes and modifications to these embodiments may be made by those skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications are within the scope of the invention.
Claims (8)
1. A3D global free-form deformation method, characterized in that the 3D global free-form deformation method comprises:
acquiring a 3D image;
matching the 3D image with a pre-stored image library, wherein each pre-stored image in the pre-stored image library is provided with a tensor model for adjusting the shape of image points of the pre-stored image, and the tensor model is a functional expression which is arranged on the pre-stored image and represents the relationship between the image points;
generating a 3D image with a tensor model arranged on an image point by utilizing the pre-stored tensor model on the image through an artificial intelligence deep learning algorithm;
and the image points on the 3D image provided with the tensor model on the image points are deformed through the tensor model.
2. The 3D global free-form deformation method of claim 1, wherein the 3D global free-form deformation method comprises:
and setting a function formula between image points on the 3D image by utilizing the pre-stored function formula on the image through an artificial intelligence deep learning algorithm.
3. The 3D global free-form deformation method according to claim 2, wherein each pre-stored image in the pre-stored image library is divided into a plurality of regions, each region having a function expression indicating a relationship between image points in the same region, the 3D global free-form deformation method comprises:
dividing regions on the 3D image by using the region positions on the pre-stored image through an artificial intelligence deep learning algorithm;
and for a target area on the 3D image, setting a function between image points in the target area on the 3D image by using the pre-stored image function through an artificial intelligence deep learning algorithm.
4. The 3D global free-form deformation method of claim 3, wherein the 3D global free-form deformation method comprises:
for a target pre-stored image in a pre-stored image library, acquiring a function formula between adjacent image points in the target pre-stored image, wherein the function formula is a polynomial function;
and obtaining a plurality of parting lines passing through adjacent image points through artificial intelligence deep learning, calculating the sum of times of the highest-order item of a polynomial function between all the adjacent image points passing through the parting lines, and dividing the region of the target pre-stored image by the parting lines with the sum of times lower than a preset value.
5. The 3D global free deformation method of claim 1, wherein the generating the 3D image with the tensor model on the image points by the artificial intelligence deep learning algorithm comprises:
matching the 3D image with a target image in the prestored image library;
adjusting the space shape of the target image according to the space shape of the 3D image through an artificial intelligence deep learning algorithm;
and taking the target image with the adjusted space shape as a 3D image with a tensor model on the image point.
6. The 3D global free-form deformation method of claim 5, wherein the adjusting the spatial shape of the target image according to the spatial shape of the 3D image through an artificial intelligence deep learning algorithm comprises:
overlapping the 3D image and the target image to obtain the distance from an image point on the target image to the 3D image;
acquiring the image point with the maximum distance as a control point, and moving the control point to the direction of the 3D image by the target length;
and moving the surrounding control points around the control point to the direction of the 3D image by using a tensor model of the target image to calculate the length, wherein the size of the calculated length of each surrounding control point is in inverse proportion to the distance from the surrounding control point to the control point, and the calculated length is less than the target length.
7. The 3D global free deformation method according to claim 1, wherein the deforming the image point on the 3D image with the tensor model on the image point by the tensor model comprises:
acquiring an adjusting instruction for adjusting a target image point on the 3D image;
moving the target image point to a target direction by an instruction length according to an adjusting instruction;
and moving the peripheral image points around the target image point to the target direction by utilizing the tensor model of the 3D image to calculate the length, wherein the size of the calculated length of each peripheral image point is in inverse proportion to the distance from the peripheral image point to the target image point, and the calculated length is less than the target length.
8. 3D global free deformation apparatus, characterized in that the 3D global free deformation apparatus is used to implement the 3D global free deformation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811403638.4A CN109636926B (en) | 2018-11-23 | 2018-11-23 | 3D global free deformation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811403638.4A CN109636926B (en) | 2018-11-23 | 2018-11-23 | 3D global free deformation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109636926A CN109636926A (en) | 2019-04-16 |
| CN109636926B true CN109636926B (en) | 2023-02-07 |
Family
ID=66068883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811403638.4A Active CN109636926B (en) | 2018-11-23 | 2018-11-23 | 3D global free deformation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109636926B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111161399B (en) * | 2019-12-10 | 2024-04-19 | 上海青燕和示科技有限公司 | Data processing method and assembly for generating three-dimensional model based on two-dimensional image |
| CN111862046B (en) * | 2020-07-21 | 2023-11-17 | 江苏省人民医院（南京医科大学第一附属医院） | Catheter position discrimination system and method in heart coronary wave silhouette |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096377A (en) * | 2014-05-14 | 2015-11-25 | 华为技术有限公司 | Image processing method and apparatus |
| CN107067393A (en) * | 2017-04-14 | 2017-08-18 | 大连理工大学 | A kind of three-dimensional medical image segmentation method based on user mutual and shape prior knowledge |
| CN108182699A (en) * | 2017-12-28 | 2018-06-19 | 北京天睿空间科技股份有限公司 | Three-dimensional registration method based on two dimensional image local deformation |
| CN108682050A (en) * | 2018-08-16 | 2018-10-19 | Oppo广东移动通信有限公司 | U.S. face method and apparatus based on threedimensional model |
-
2018
- 2018-11-23 CN CN201811403638.4A patent/CN109636926B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096377A (en) * | 2014-05-14 | 2015-11-25 | 华为技术有限公司 | Image processing method and apparatus |
| CN107067393A (en) * | 2017-04-14 | 2017-08-18 | 大连理工大学 | A kind of three-dimensional medical image segmentation method based on user mutual and shape prior knowledge |
| CN108182699A (en) * | 2017-12-28 | 2018-06-19 | 北京天睿空间科技股份有限公司 | Three-dimensional registration method based on two dimensional image local deformation |
| CN108682050A (en) * | 2018-08-16 | 2018-10-19 | Oppo广东移动通信有限公司 | U.S. face method and apparatus based on threedimensional model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109636926A (en) | 2019-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11495002B2 (en) | Systems and methods for determining the scale of human anatomy from images | |
| CN106162203B (en) | Panoramic video playback method, player and wear-type virtual reality device | |
| CN108124509B (en) | Image display method, wearable intelligent device and storage medium | |
| RU2565855C1 (en) | Image capturing device, method of controlling said device and programme | |
| US20130188019A1 (en) | System and Method for Three Dimensional Imaging | |
| CN108600729B (en) | Dynamic 3D model generation device and image generation method | |
| CN109636926B (en) | 3D global free deformation method and device | |
| CN108391116B (en) | Whole body scanning device and method based on 3D imaging technology | |
| CN109657559B (en) | Point cloud depth perception coding engine device | |
| CN111161399B (en) | Data processing method and assembly for generating three-dimensional model based on two-dimensional image | |
| CN116980579B (en) | Image stereoscopic imaging method based on image blurring and related device | |
| CN108513122B (en) | Model adjusting method and model generating device based on 3D imaging technology | |
| CN109657702B (en) | 3D depth semantic perception method and device | |
| CN117133310A (en) | Audio drive video generation method and device, storage medium and electronic equipment | |
| KR101873161B1 (en) | Method and apparatus for providing personal 3-dimensional image using convergence matching algorithm | |
| CN111629194B (en) | Method and system for converting panoramic video into 6DOF video based on neural network | |
| CN109448066B (en) | 3D data polar compression method and device | |
| CN109379511B (en) | 3D data security encryption algorithm and device | |
| CN109272453B (en) | Modeling device and positioning method based on 3D camera | |
| CN109218703B (en) | Data processing device and method based on 3D camera | |
| CN109089105B (en) | Model generation device and method based on depth perception coding | |
| CN109151437B (en) | Whole body modeling device and method based on 3D camera | |
| CN109348208B (en) | Perception code acquisition device and method based on 3D camera | |
| CN109218699B (en) | Image processing device and method based on 3D camera | |
| CN109379509B (en) | Point cloud encryption compression engine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230406 Address after: 518000 1101-g1, BIC science and technology building, No. 9, scientific research road, Maling community, Yuehai street, Nanshan District, Shenzhen, Guangdong Province Patentee after: Shenzhen Yujing Information Technology Co.,Ltd. Address before: 201703 No.206, building 1, no.3938 Huqingping Road, Qingpu District, Shanghai Patentee before: UNRE (SHANGHAI) INFORMATION TECHNOLOGY Co.,Ltd. |