KR101698133B1 - Customized human body modeling apparatus and method - Google Patents
Customized human body modeling apparatus and method Download PDFInfo
- Publication number
- KR101698133B1 KR101698133B1 KR1020150084242A KR20150084242A KR101698133B1 KR 101698133 B1 KR101698133 B1 KR 101698133B1 KR 1020150084242 A KR1020150084242 A KR 1020150084242A KR 20150084242 A KR20150084242 A KR 20150084242A KR 101698133 B1 KR101698133 B1 KR 101698133B1
- Authority
- KR
- South Korea
- Prior art keywords
- model
- human body
- template model
- statistical template
- scan data
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T13/00—Animation
- G06T13/20—3D [Three Dimensional] animation
- G06T13/40—3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
A personal body modeling apparatus and method is disclosed. The personalized body modeling apparatus includes a scanning unit that scans a human body and generates scan data for the human body; and a controller that modifies a statistical template model associated with the human body based on the scan data to generate a customized human shape model And a processor for forming the processor.
Description
Embodiments of the present invention relate to techniques for forming a custom body shape model for a human body based on a statistical template model.
In recent years, the design of a specific person's body shape has been highlighted in various industrial fields such as human body, clothing, game, and automobile. This is the main purpose of improving the convenience and safety by considering the shape of a specific person in the design process of the product, and providing more useful products to the individual.
The conventional human shape generating apparatus can generate a human body shape model of a human body realistically and specifically, but it requires high cost of scanning equipment that supports precise scanning, thus requiring a high cost.
In addition, the human shape generating apparatus can increase the completeness of the noise and the shape surface through a plurality of short-range scans using a low-cost scanning apparatus, but it is considerably cumbersome due to a need to scan a stopped scan object several hundred times, The scan data is inaccurate due to the motion of the scan object, and therefore the quality of the human body shape generated based on the scan data is inevitably lowered.
The present invention relates to a method and apparatus for generating scan data for a human body through a Kinect and modifying a statistical template model associated with the human body based on the scan data to form a customized human shape model for the human body, A statistical template model reflecting the shape of a human body belonging to a population is transformed into a target of scan data and thus a high quality customized human shape model is provided without repeated scans on the human body .
According to the present invention, a size (or a ratio) of a statistical template model associated with a human body is primarily deformed based on joint information extracted from scan data (e.g., joint position and joint length) The object of the present invention is to provide a customized human shape model with a higher quality by deforming the curved surface of the statistical template model so that the curved surfaces of the statistical template model coincide with each other.
According to another aspect of the present invention, there is provided a personal body modeling apparatus for a human body, comprising a scan unit for scanning a human body and generating scan data for the human body, and a controller for modifying a statistical template model associated with the human body based on the scan data, To form a customized human shape model.
The customized human body modeling apparatus may further include a model construction unit that implements the statistical template model using three-dimensional scan data for each human body included in the population to which the human body belongs.
The processor may modify the statistical template model so that the difference between the curved surface of the scan model formed using the scan data and the curved surface of the statistical template model satisfies a set range.
Wherein the statistical template model includes a first function for determining the extent to which the statistical template model is maintained and reflected in the customized human shape model, a second function for curve matching between the scan model and the statistical template model, The statistical template model can be modified by using at least one function among the third function relating to joint matching between models.
Wherein the processor assigns weights to the first function, the second function, and the third function, respectively, so that the curved surface of the scan model and the curved surface of the statistical template model coincide with each other, , And the sum of the three functions, the statistical template model can be modified.
The scan unit may include n (n is a natural number equal to or greater than 3) number of kinect, which are installed at different positions or directions with respect to the human body and generate scan data for each part of the human body.
The processor may combine the scan data generated in each of the n keynets with the information about the n kinks into consideration, and modify the statistical template model using the combined scan data.
According to another aspect of the present invention, there is provided a method of modeling a human body, the method comprising: scanning a human body to generate scan data for the human body; modifying a statistical template model associated with the human body based on the scan data; And forming a customized human shape model.
According to the embodiment of the present invention, scan data for a human body is generated through a key knot, and a statistical template model associated with the human body is modified based on the scan data to form a custom human shape model for the human body Thus, by using a cheap Kinect, a statistical template model reflecting the shape of the human body belonging to the population is transformed by targeting the scan data, so that a high-quality customized human shape model Can be provided.
According to the embodiment of the present invention, the size (or ratio) of the statistical template model associated with the human body is primarily modified based on the joint information extracted from the scan data (for example, the joint position and the joint length) The curved surface of the statistical template model is secondarily deformed so that the curved surface of the model coincides with the curved surface of the statistical template model to provide a customized human shape model of improved quality.
1 is a diagram illustrating a configuration of a custom body modeling apparatus according to an embodiment of the present invention.
2 is a view for explaining an example of a human body modeling process in a customized human body modeling apparatus according to an embodiment of the present invention.
3 is a diagram for explaining an example of a landmark and a statistical template model used in a customized human body modeling apparatus according to an embodiment of the present invention.
FIG. 4 is a view showing an example of joint information extracted from scan data in a customized human body modeling apparatus according to an embodiment of the present invention.
FIG. 5 is a view showing an example of deformation from a scan model and a statistical template model until a customized human shape model is formed by a customized human body modeling apparatus according to an embodiment of the present invention.
FIG. 6 is a view for explaining an example of generating scan data in the customized human body modeling apparatus according to an embodiment of the present invention.
7 is a flowchart illustrating a method of customizing human body modeling according to an embodiment of the present invention.
Hereinafter, an apparatus and method for customizing human body modeling according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.
1 is a diagram illustrating a configuration of a custom body modeling apparatus according to an embodiment of the present invention.
Referring to FIG. 1, a customized human
The
The
The
In addition, the
Where E can be a statistical objective function.
Is a data error function that calculates the sum of the squared distances between points so that the curved surface of the statistical template model is as close as possible to the sample model and can be expressed by Equation (2). Also, The ≪ / RTI >
here,
Of the statistical template model Third, A corresponding 4x4 affine transform matrix, In the target model A matching point corresponding to < RTI ID = 0.0 & Is a weight for controlling the influence of the data error function.Is a function of the deformation error. It is a function of squaring the difference between the transformation of a point on the statistical template model and the transformation of its neighboring points, and can be expressed by Equation (3). Also, The ≪ / RTI >
Also,
Is a landmark error function that calculates the difference between landmarks between two curved surfaces so that a particular portion of the statistical template model and the corresponding curved surface portion of the sample model are coincident, . Also, The ≪ / RTI >
here,
Is the number of landmarks, Of the target model Th landmark point, Is an index of a point corresponding to each landmark on the statistical template model.In addition, the
The
The
At this time, the
At the time of the transformation, the
Specifically, the
Thereafter, the
At this time, the
Where E can be the target objective function.
Is a registration error function, which means to maintain the shape of the existing statistical template model at the time of transformation, and can be expressed by Equation (6). Also, The ≪ / RTI >
here,
Of the statistical template model Third, The Th < th > point, Is a corresponding 4x4 affine transform matrix.Is a data error function that matches the curved surface of the statistical template model as close as possible to the curved surface of the scan model (target model) and can be expressed by Equation (7). Also, The ≪ / RTI >
here,
The number of points constituting the model, Is a weight for adjusting the influence of data error. Also, Of the template model Third, In the scan model (target model) Lt; / RTI >Is a connectivity error function that matches the joint positions between the statistical template model and the scan model (target model), and can be expressed by Equation (8). Also, The ≪ / RTI >
here,
Is the number of joints, Is the coordinates of the joint point of the statistical template model, Is the coordinates of the joint point of the scan model (target model).The
The personalized human
2 is a view for explaining an example of a human body modeling process in a customized human body modeling apparatus according to an embodiment of the present invention. FIG. 3 is a view for explaining an example of landmark and statistical template model implementation, FIG. 4 is an illustration of an example of joint information extracted from scan data, FIG. 5 is a diagram Fig.
Referring to FIG. 2, the customized human body modeling apparatus according to an embodiment of the present invention acquires three-dimensional scan data for each human body included in the population of '20 to 39 year old male' The
Here, the customized human body modeling apparatus is a parameter commonly used for each sample model in order to statistically analyze the human body shape. For example, a landmark (for example, 64 landmarks) can be specified. The landmark specified in the sample model can be utilized in the surface fitting process for statistical template model implementation.
Also, as shown in FIG. 3 (b), the customized human body modeling apparatus can generate a statistical template model by matching the respective sample models based on the landmarks specified in the individual sample models, Is divided into a plurality of regions, and a statistical template model including joint information can be implemented through the human segmentation step.
The customized human body modeling device is a specific human body that scans the human body of a '25 year old male' to generate
Specifically, the customized human body modeling device extracts joint information including at least one of the position of the joint and the length of the joint from the
Thereafter, the customized human body modeling apparatus forms a
As a result, as shown in FIG. 5, the customized human body modeling apparatus forms a customized human body shape model for the human body using a scan model and a statistical template model for the human body, And the line (or the surface) of the modified statistical template model is modified secondarily to reflect the characteristics of the scan model so as to form a highly matched customized human shape model .
6 is a view for explaining an example of a human body scan in a customized human body modeling apparatus according to an embodiment of the present invention.
Referring to FIG. 6, the customized human body modeling apparatus can scan a specific human body through the scan unit and generate scan data for the human body. At this time, the customized human body modeling apparatus can scan the human body through three first, second, and
The first, second, and
The customized human body modeling device transmits the generated scan data through the first, second and
7 is a flowchart illustrating a method of customizing human body modeling according to an embodiment of the present invention.
Referring to FIG. 7, in
In
In step 705, the customized human body modeling device can modify the statistical template model associated with the human body based on the scan data for the human body, thereby forming a customized human shape model for the human body. At this time, the customized human body modeling apparatus can identify a population to which the human body belongs, detect a statistical template model corresponding to the identified population from the database, and use the statistical template model associated with the human body.
In the modification of the statistical template model, the customized human body modeling device extracts joint information including at least one of the position of the joint and the length of the joint from the scan data, and based on the extracted joint information, The size or proportion of the model can be modified primarily. At this time, if the customized human body modeling apparatus generates scan data by the n number of keynets, the individual human body modeling apparatus generates each scan data by using information on the n number of keynets (e.g., installation position, direction, And the like), and transforms the statistical template model using the combined scan data.
Thereafter, the customized human body modeling device can secondarily modify the curved surface of the statistical template model so that the difference between the curved surface of the scan model formed using the scan data and the curved surface of the statistical template model satisfies the set range.
In this case, the customized human body modeling apparatus may further include a first function for determining the extent to which the statistical template model is maintained and reflected in the customized human shape model, a second function for curve matching between the scan model and the statistical template model, And a third function relating to joint matching between the statistical template model, may be used to transform the statistical template model. Herein, the customized human body modeling device assigns weight values to the first function, the second function, and the third function so that the curved surface of the scan model and the curved surface of the statistical template model coincide with each other, The statistical template model can be modified using a target objective function that is a sum of the first, second, and third functions.
An embodiment of the present invention is characterized by generating scan data for a human body through a Kinect and modifying a statistical template model associated with the human body based on the scan data to form a customized human shape model for the human body, By reducing the cost of using cheap Kinect and transforming the statistical template model that reflects the shape of the human body belonging to the population as the target of the scan data, it provides a high quality custom body shape model without repetitive scans on the human body can do.
In the embodiment of the present invention, the size (or ratio) of the statistical template model associated with the human body is primarily modified based on the joint information extracted from the scan data (for example, the joint position and the inter-joint length) The curved surface of the statistical template model is secondarily deformed so that the curved surface and the curved surface of the statistical template model coincide with each other, thereby providing a customized humanoid shape model of improved quality.
The method according to an embodiment may be implemented in the form of a program instruction that may be executed through various computer means and stored in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions stored on the medium may be those specially designed and constructed for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable storage media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magneto-optical media such as floppy disks; Includes hardware devices specifically configured to store and execute program instructions such as magneto-optical media and ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.
While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.
Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.
100: Customized body modeling device
101:
103:
105: Processor
107: Database
Claims (15)
A processor for transforming a statistical template model associated with the human body based on the scan data to form a customized human shape model for the human body,
Lt; / RTI >
The scan unit
And n (n is a natural number equal to or greater than 3) number of kinect, which are provided at different positions or directions with respect to the human body and generate scan data for each part of the human body,
The processor
The statistical template model is used to supplement the portion of the human body not included in the scan data to be included in the scan data,
Modifying the statistical template model so that a difference between a curved surface of a scan model formed using the scan data and a curved surface of the statistical template model satisfies a set range,
The statistical template model
Wherein the template model is detected from a database corresponding to a population to which the human body belongs.
And a model construction unit for implementing the statistical template model using three-dimensional scan data for each human body included in the population to which the human body belongs,
Further comprising:
The processor comprising:
Extracting joint information including at least one of the position of the joint and the length of the joint from the scan data and modifying the size or the ratio of the statistical template model based on the extracted joint information
Customized body modeling device.
The processor comprising:
Wherein the statistical template model includes a first function for determining the degree to which the statistical template model is maintained and reflected in the customized human shape model,
A second function relating to the surface matching between the scan model and the statistical template model,
And a third function relating to joint matching between the scan model and the statistical template model to transform the statistical template model
Customized body modeling device.
The processor comprising:
The first function, the second function, and the third function so that the curved surface of the scan model and the curved surface of the statistical template model coincide with each other, and the weighted first, second, and third functions Of the statistical template model using the target objective function,
Customized body modeling device.
The processor comprising:
Combining the scan data generated in each of the n keykits in consideration of the information on the n kinks and modifying the statistical template model using the combined scan data
Customized body modeling device.
Modifying a statistical template model associated with the human body based on the scan data to form a customized human shape model for the human body
Lt; / RTI >
The step of generating the scan data
Generating scan data for each part of the human body through n (n is a natural number equal to or larger than 3) kinks provided at different positions or directions with respect to the human body,
Wherein the step of forming the customized human-
Supplementing the portion of the human body not included in the scan data to be included in the scan data using the statistical template model; And
Transforming the statistical template model so that a difference between a curved surface of a scan model formed using the scan data and a curved surface of the statistical template model satisfies a set range
Lt; / RTI >
The statistical template model
Wherein the template model is detected from a database corresponding to the population to which the human body belongs.
Implementing the statistical template model using three-dimensional scan data for each human body included in the population to which the human body belongs,
Further comprising the steps of:
Wherein the step of forming the customized human-
Extracting joint information including at least one of a position of the joint and a length of the joint from the scan data; And
Modifying a size or a ratio of the statistical template model based on the extracted joint information
Wherein the method comprises the steps of:
Wherein transforming the statistical template model comprises:
Wherein the statistical template model includes a first function for determining the degree to which the statistical template model is maintained and reflected in the customized human shape model,
A second function relating to the surface matching between the scan model and the statistical template model,
Transforming the statistical template model using at least one function of a third function relating to joint matching between the scan model and the statistical template model
Wherein the method comprises the steps of:
Wherein transforming the statistical template model comprises:
Assigning weights to the first function, the second function, and the third function, respectively, so that the curved surface of the scan model and the curved surface of the statistical template model coincide with each other; And
Modifying the statistical template model using a target objective function consisting of the sum of the weighted first, second and third functions
Further comprising the steps of:
Wherein the step of forming the customized human-
Combining the scan data generated in each of the n kinks by considering information on the n kinks, and modifying the statistical template model using the combined scan data
Wherein the method comprises the steps of:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150084242A KR101698133B1 (en) | 2015-06-15 | 2015-06-15 | Customized human body modeling apparatus and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020150084242A KR101698133B1 (en) | 2015-06-15 | 2015-06-15 | Customized human body modeling apparatus and method |
Publications (2)
Publication Number | Publication Date |
---|---|
KR20160147466A KR20160147466A (en) | 2016-12-23 |
KR101698133B1 true KR101698133B1 (en) | 2017-01-19 |
Family
ID=57736052
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
KR1020150084242A KR101698133B1 (en) | 2015-06-15 | 2015-06-15 | Customized human body modeling apparatus and method |
Country Status (1)
Country | Link |
---|---|
KR (1) | KR101698133B1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102655987B1 (en) * | 2018-11-21 | 2024-04-11 | 한국전자통신연구원 | Apparatus and method for generating 3d avatar |
CN110321090B (en) * | 2019-07-12 | 2024-02-20 | 广东工业大学 | Protective clothing manufacturing device and method |
KR102402031B1 (en) | 2020-11-27 | 2022-05-27 | 임대순 | System for manufacturing customized clothing using shape memory material |
CN116269455B (en) * | 2023-03-20 | 2023-12-12 | 瑞石心禾(河北)医疗科技有限公司 | Detection method and system for automatically acquiring human body contour in SPECT (single photon emission computed tomography) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101466123B1 (en) * | 2014-05-27 | 2014-11-28 | 한양대학교 산학협력단 | Parametric modeling method of human body shape using statistical analysis |
-
2015
- 2015-06-15 KR KR1020150084242A patent/KR101698133B1/en active IP Right Grant
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101466123B1 (en) * | 2014-05-27 | 2014-11-28 | 한양대학교 산학협력단 | Parametric modeling method of human body shape using statistical analysis |
Non-Patent Citations (1)
Title |
---|
Jing Tong ET AL., Scanning 3D Full Human Bodies using Kinects, IEEE, 2012.03.09., 643-650P* |
Also Published As
Publication number | Publication date |
---|---|
KR20160147466A (en) | 2016-12-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102647351B1 (en) | Modeling method and modeling apparatus using 3d point cloud | |
CN106228162B (en) | A kind of quick object identification method of mobile robot based on deep learning | |
KR101997500B1 (en) | Method and apparatus for generating personalized 3d face model | |
KR101994316B1 (en) | Device and method for object pose estimation | |
US9978119B2 (en) | Method for automatic facial impression transformation, recording medium and device for performing the method | |
KR101698133B1 (en) | Customized human body modeling apparatus and method | |
JP4950787B2 (en) | Image processing apparatus and method | |
CN109598781B (en) | Method for acquiring pseudo 3D frame from 2D bounding frame by regression analysis, learning apparatus and testing apparatus using the same | |
US20080285801A1 (en) | Visual Tracking Eye Glasses In Visual Head And Eye Tracking Systems | |
US10839526B2 (en) | Information processing device, information processing method, and recording medium | |
KR20170000748A (en) | Method and apparatus for face recognition | |
CN104346608A (en) | Sparse depth map densing method and device | |
CN110148179A (en) | A kind of training is used to estimate the neural net model method, device and medium of image parallactic figure | |
KR101971791B1 (en) | System for providing of augmented reality in indoor space using visible light communication and the method thereof | |
JP6409433B2 (en) | Image generation apparatus, image detection system, and image generation method | |
KR102107021B1 (en) | Method, system and non-transitory computer-readable recording medium for providing an identification of data | |
CN113168729B (en) | 3D shape matching method and device based on local reference coordinate system | |
JPWO2018207282A1 (en) | Object recognition method, apparatus, system, program | |
JP6351350B2 (en) | 3D model retrieval system and 3D model retrieval method | |
KR20180078115A (en) | Method and apparatus for processing image | |
JP7224592B2 (en) | Information processing device, information processing method, and program | |
JP2015219868A (en) | Information processor, information processing method and program | |
JP6079076B2 (en) | Object tracking device and object tracking method | |
JP2009288917A (en) | Information processor, information processing method and program | |
JP6396353B2 (en) | Determination apparatus and determination method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
A201 | Request for examination | ||
E902 | Notification of reason for refusal | ||
AMND | Amendment | ||
E601 | Decision to refuse application | ||
AMND | Amendment | ||
X701 | Decision to grant (after re-examination) |