CN110074788B - Body data acquisition method and device based on machine learning - Google Patents
Body data acquisition method and device based on machine learning Download PDFInfo
- Publication number
- CN110074788B CN110074788B CN201910316421.8A CN201910316421A CN110074788B CN 110074788 B CN110074788 B CN 110074788B CN 201910316421 A CN201910316421 A CN 201910316421A CN 110074788 B CN110074788 B CN 110074788B
- Authority
- CN
- China
- Prior art keywords
- human body
- model
- measurer
- data acquisition
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1075—Measuring physical dimensions, e.g. size of the entire body or parts thereof for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1077—Measuring of profiles
- A61B5/1078—Measuring of profiles by moulding
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/107—Measuring physical dimensions, e.g. size of the entire body or parts thereof
- A61B5/1079—Measuring physical dimensions, e.g. size of the entire body or parts thereof using optical or photographic means
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G19/00—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups
- G01G19/44—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing persons
- G01G19/50—Weighing apparatus or methods adapted for special purposes not provided for in the preceding groups for weighing persons having additional measuring devices, e.g. for height
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/04—Texture mapping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/08—Indexing scheme for image data processing or generation, in general involving all processing steps from image acquisition to 3D model generation
Abstract
The invention discloses a body data acquisition method and a body data acquisition device based on machine learning, wherein the method comprises the following steps: acquiring an image and weight of a measurer; combining the image and the weight of the measurer with the human body 3D model, and outputting the human body 3D model with the chartlet of the measurer and the body data information. The invention also provides a body data acquisition device based on machine learning, which comprises a host and a weight scale; the host comprises a display, a camera and an operating console; the operation panel includes identification device, controller and printer, display, camera, identification device and printer respectively with the controller electricity is connected, the weighing scale with controller communication connection. The invention does not make specific requirements on clothes of people, the obtained body data is not influenced by the clothes, the body state and posture of the people can be accurately presumed under the condition of wearing daily clothes, and the convenience of measuring the body is greatly improved.
Description
Technical Field
The invention belongs to the technical field of human body data measurement, and particularly relates to a method and a device for acquiring body data based on machine learning.
Background
The traditional method for acquiring the body shape data of the human body through manual measurement requires that a measuring person must be capable of contacting the measured person, remote measurement cannot be achieved, and various errors are possible in the traditional measuring method. With the continuous development of the 3D technology, the human body 3D scanner is also developed, but the whole body scanning technology is used, the human body is required to be fully exposed, a human body model is obtained, the equipment cost is high, the size is large, remote measurement cannot be realized, and the high price is difficult to widely popularize in daily life. The existing instrument for measuring the human body depends on depth equipment, the function detection is incomplete, and certain deviation can exist due to shielding of clothes and the like.
In the current field, there are mainly two types of human body three-dimensional modeling methods: the first type is that a simple skeleton model map of a human body is obtained by looking at a group of joints, such as a neck, such as an elbow, the data does not contain body dimension information of an analysis object, and the analysis information is very limited. The second type is to use a plurality of groups of cameras to shoot a plurality of 3D models by means of a 3D scanning technology, and the model obtained by the method cannot exclude the influence of clothes and hairstyle on the body. The current method for three-dimensional modeling of human bodies has very limited analysis information, cannot eliminate the influence of clothes and hairstyles on body models, and has very limited application fields of the obtained models. A model obtained by splicing the pictures taken by the depth camera in the multi-angle shooting method does not have body scale data and cannot provide a basis for 3D perception.
Disclosure of Invention
The invention provides a body data acquisition method and device based on machine learning, and solves the problems of inaccurate body data detection result, complex device structure, inconvenience in use and the like in the prior art.
In order to realize the purpose of the invention, the invention adopts the following technical scheme to realize:
a method of machine learning-based body data acquisition, comprising the steps of:
acquiring an image and weight of a measurer;
combining the image and the weight of the measurer with the human body 3D model, and outputting the human body 3D model with the chartlet of the measurer and the body data information.
Further, the method for generating the human body 3D model comprises:
randomly selecting training crowds, and shooting to obtain training pictures;
dividing key parts of a human body on a training picture into semantic regions;
manually marking key points on the key parts, and manually determining the natural direction of the key parts to obtain training data;
and combining the training data with the standard human body model template, and training to obtain the human body 3D model.
Further, the key parts of the human body are as follows: the left head, the right head, the neck, the left upper arm, the left lower arm, the right upper arm, the trunk, the left hand, the right hand, the left thigh, the right thigh, the left shank, the right shank, the left foot and the right foot, wherein the upper arm, the lower arm, the thigh and the shank are composed of a front part and a back part.
Further, the artificially labeled key points are key points on the edge of the body and key parts of the human body.
Further, the standard human body model template is as follows:
wherein each one
The model template is corresponding to the vertexes of a 3-dimensional space human model, m is the total number of vertexes of a standard human model template, X represents the set of vertexes of the standard human model template, the standard human model template is a standard 3D grid formed by triangles, and the number of the triangles is 5500-6500.
Further, the surveyor's image is a topological variation of a 3D model of the human body and a vector mapping, a bi-directional mapping ψ, from 3D coordinates to 2D space
Thereby to obtain
Psi: represents a mapping relationship, uj: 2D space coordinate, xj: and the human body 3D model coordinates represent a vertex in the 3D mesh, and UV coordinates of the human body 3D model and the chartlet of the measurer are obtained.
Furthermore, the measurer image is used as a map, the map is colored for the measurer human body 3D model according to the UV coordinates of the map, and the measurer human body 3D model with the map is output.
A body data acquisition device based on machine learning comprises a host and a body weight scale; the host comprises a display, a camera and an operating console; the operation panel includes identification device, controller and printer, display, camera, identification device and printer respectively with the controller electricity is connected, the weighing scale with controller communication connection.
Further, the display is arranged at the upper part of the host, the camera is arranged at the middle part of the host, and the operating platform is arranged at the lower part of the host; the identity recognition module comprises a card reader and a scanner.
Further, the display is a touch display screen or a projector.
The invention has the advantages and beneficial effects that: the method and the device for acquiring the body data based on the machine learning can eliminate the influence of loose clothes and hairstyle, maximally approach the body shape of a human body under the condition that the human body does not wear clothes, and accurately separate the outline of the human body under a complex background. According to the invention, the human body 3D model can be obtained corresponding to the real shape of the human body without a depth-of-field camera or a plurality of groups of cameras, and a wide application scene is provided for various industries, such as clothes, health and the like.
The invention provides a method for analyzing a full-body picture of a human body through a deep neural network to obtain an accurate human body three-dimensional model. The principle of the method can be generally applied to other fields of 2D to 3D conversion, and face modeling, ear modeling and foot modeling are carried out in a finer coordinate segmentation field.
The device of the invention provides two authentication modes, which not only facilitates data acquisition, but also is beneficial to privacy protection; the device has more perfect function detection by using a body data acquisition method based on machine learning, provides accurate information of body data such as height, shoulder width, arm length, leg length and the like of a human body by combining data such as weight and the like, and can provide a proper clothes model of a measurer; the acquired data can be printed in real time, and the printing is convenient and quick. The device has simple structure, easy operation and lower cost, is convenient for large-scale popularization and use, and can be widely applied to fitness crowds, fitness mechanisms, garment customization mechanisms and the like.
Drawings
FIG. 1 is a flow chart of a method of machine learning based body data acquisition;
FIG. 2 is a schematic view of marking of a main body part; wherein, a is an original picture, and b is a picture of a head and a trunk which are manually marked;
fig. 3 is a schematic diagram of outputting a 3D model of a human body of a measurer with a map by using a human image as an input.
FIG. 4 is a schematic diagram of a body data acquisition device based on machine learning;
FIG. 5 is a side view of a body data acquisition device host based on machine learning;
wherein, 1 is a host computer, 11 is a display, 12 is a camera, 13 is an operation desk, 131 is a printer, 132 is a card reader, 133 is a scanner, and 2 is a weighing scale.
Detailed Description
The technical solution of the present invention will be described in further detail with reference to specific embodiments.
Example 1
The invention provides a body data acquisition method based on machine learning, which utilizes a marking method of training data to mark quickly and divide human body semanteme, and can accurately reconstruct an algorithm model of a body under the scenes of complex background and different distances and provide body data information. Fig. 1 is a flow chart of a method of machine learning based body data acquisition, the method comprising the steps of:
(1) and acquiring the image and the weight of the measurer.
Specifically, a monocular camera is used to take an image of the measurer, and the weight of the measurer is acquired.
(2) And combining the image and the weight of the measurer with the human body 3D model, and outputting the human body 3D model with the chartlet of the measurer and body data information.
The generation method of the human body 3D model comprises the following steps:
a. training crowds are randomly selected, and training pictures are obtained through shooting.
Specifically, a training crowd is randomly selected, photographed in various scenes, and the photographed image is used as a training picture.
b. And dividing key parts of the human body on the training picture into semantic regions.
Specifically, key parts of a human body on a training picture are divided into 24 semantic regions, wherein the key parts comprise a left head, a right head, a neck, a left upper arm, a right upper arm, a left lower arm, a right lower arm, a trunk, a left hand, a right hand, a left thigh, a right thigh, a left calf, a right calf, a left foot and a right foot; wherein, the upper arm, the lower arm, the thigh and the shank are composed of a front part and a back part.
c. And manually marking key points on the key parts, and manually determining the natural direction of the key parts to obtain training data.
Specifically, as shown in fig. 2, on the training picture, the edge of the human body (not the edge of the clothing) and the key part of the human body are manually marked, and then key points are selected for marking, and meanwhile, according to the joint position in the key part, the natural direction of the part, such as the angle separating the left upper arm and the human body, is determined, and a large amount of training data is accumulated.
d. And combining the training data with the standard human body model template, and training to obtain the human body 3D model.
Wherein, the standard human body model template is as follows:
wherein each one
All correspond to the vertexes of a 3-dimensional space human body model, m is the total number of the vertexes of the standard human body model template, and X represents the set of the vertexes of the standard human body model template(ii) a The standard mannequin template is a standard 3D mesh consisting of 6000 triangles.
The measurer image is combined with the generated human body 3D model, so that the human body 3D model with the chartlet of the measurer and the body data information can be output, and the method specifically comprises the following steps:
when defining the machine learning model, the input is the image of the measurer, the feature captured from the image can realize a group of mapping, the point-to-point mapping of 2D and 3D is completed, and from the group of mapping, the 3D model of the human body of the measurer, namely the depth model, and the UV coordinate of the mapping can be obtained.
Inputting the image of the measurer obtained in the step (5), wherein the image is in an RGB format and can be a human body image in any posture; the surveyor's body is the topological variation and vector mapping of the human body's 3D model. Bidirectional mapping psi from 3D coordinates to 2D space
Thereby to obtain
Wherein ψ: represents a mapping relationship, uj: 2D space coordinate, x, obtained from a picture of a target human body to be measuredj: human 3D model coordinates, representing one vertex in 3D mseh. The 3D human body model template is composed of a plurality of triangles, the stretching, rotation and displacement of the triangles correspond to the changes of human bodies, and a group of topological vector mappings are extracted to represent the changes.
And directly using the picture of the human body of the measurer as a map, coloring the map for the 3D model of the human body of the measurer according to the UV coordinate of the map, and outputting the 3D model of the human body of the measurer with the map, wherein the figure is shown in figure 3. In the step, mapping is carried out on the completed 3D model, UV is coordinates of the mapping, the corresponding relation between each pixel on the picture and the 3D is required to be determined, and U and V can be trained independently, so that the complexity of calculation is reduced.
Based on training data, the above model building is done by relying on a Convolutional Neural Network (CNN). The invention aims to map each pixel on a picture to a certain area of a target template, and certainly comprises pixels which do not belong to a human body, and the mapping can generate a null output which can be ignored. This is a typical regression calculation.
The human body 3D model generated in the invention can be recycled, as long as the picture of a measurer is input, the weight of external data is combined, the body size of the human body under various different environments is detected according to machine vision analysis, the body size is matched with the clothing model in the database, and the clothing model suitable for the measurer is output, so that the method is the most rapid, quick and low-cost body data acquisition mode.
Referring to fig. 4-5, an embodiment of the present invention further provides a body data acquiring device based on machine learning, including a host 1 and a scale 2, where the scale 2 and the host 1 are in a separate structure.
The host 1 includes a display 11, a camera 12, and an operation console 13. The display 11 is a touch display screen, the display 11 is arranged on the upper portion of the host 1, the camera 12 is arranged in the middle of the host 1, and the operating console 13 is arranged on the lower portion of the host 1. The console 13 includes an identification device, a controller, and a printer 131. The controller set up in inside the operation panel 13, the controller respectively with display 11, camera 12, identification device and printer 131 electricity are connected, personal weighing scale 2 with controller communication connection, personal weighing scale 2 through bluetooth or WIFI with personal weighing scale 2 is connected. The identity recognition device comprises a card reader 132 and a scanner 133, wherein the card reader 132 is arranged in the middle of the operating platform 13 and supports recognition of second-generation identity cards; the scanner 133 is disposed on the right side of the console 13, and can acquire the identity information of the measurer by scanning the two-dimensional code. The left side of the console 13 is provided with a printer 131 for printing the generated physical sign detection report.
When the garment model is selected by using the garment model selecting device, the second-generation identity card can be placed on the card reader 132 for identity recognition, if the identity card is inconvenient to use, particularly, people in groups need to select the garment model, such as company staff or school students, the personnel can be internally coded to form a two-dimensional code, the scanner 133 is used for scanning the two-dimensional code for identity recognition, and in addition, identity information can be input through the touch display screen for direct login. After identification or direct login, the controller can acquire personal identification information. A measurer stands on the weight scale 2 to obtain weight data, the weight scale 2 transmits the weight data to the controller through Bluetooth or WIFI, and if a measurement object knows the real weight of the measurement object, the weight value can be directly and manually input through the touch display screen; the measurer aims at the camera 12, at the moment, the touch display screen can display the image captured by the camera 12, the measurer adjusts the posture according to the prompt and the requirement of the touch display screen, and after the adjustment is completed, the camera 12 takes pictures and respectively takes a front picture and a side picture of the measurer. The controller processes the obtained image and weight information of the measurer, specifically obtains body data information such as height, shoulder width and arm length by using a method of analyzing a full-body picture of a human body through a deep neural network to obtain an accurate human body three-dimensional model in the prior art, selects a matched garment model in a database in the controller according to the body data information, transmits the garment model to the printer 131, and prints a physical sign detection report with personal identity information, including the body data information (including height, shoulder width, arm length and the like) of the measurer and a proper garment model.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (5)
1. A body data acquisition method based on machine learning is characterized in that: the method comprises the following steps:
acquiring an image and weight of a measurer; the method specifically comprises the following steps: shooting an image of a measurer by using a monocular camera, and acquiring the weight of the measurer;
generating a human body 3D model, then combining the image and the weight of the measurer with the human body 3D model, and outputting the human body 3D model of the measurer with a mapping and body data information;
the method for generating the human body 3D model comprises the following steps:
randomly selecting training crowds, and shooting to obtain training pictures;
dividing key parts of a human body on a training picture into semantic regions;
manually marking key points on the key parts, and manually determining the natural direction of the key parts to obtain training data;
combining the training data with a standard human body model template, and training to obtain a human body 3D model;
the standard human body model template is a standard 3D grid formed by triangles, and the number of the triangles is 5500-6500;
subjecting the human body 3D model to topological change and vector mapping, bidirectionally mapping psi, from 3D coordinates to 2D coordinates, thereby
，
ψ: which represents the mapping relationship between the first and second objects,
: the 2D coordinates of the two-dimensional object,
: the human body 3D model coordinates represent a vertex in the 3D grid, and UV coordinates of the human body 3D model and the chartlet of the measurer are obtained;
and the measurer image is used as a map, the map is colored for the measurer human body 3D model according to the UV coordinate of the map, and the measurer human body 3D model with the map is output.
2. The machine-learning-based physical data acquisition method according to claim 1, characterized in that: the key parts of the human body are as follows: the left head, the right head, the neck, the left upper arm, the left lower arm, the right upper arm, the trunk, the left hand, the right hand, the left thigh, the right thigh, the left shank, the right shank, the left foot and the right foot, wherein the upper arm, the lower arm, the thigh and the shank are composed of a front part and a back part.
3. The machine-learning-based physical data acquisition method according to claim 1, characterized in that: the artificially marked key points are key points on the edge of the body and key parts of the human body.
4. A body data acquisition apparatus using the body data acquisition method based on machine learning of claim 1, characterized in that: it comprises a host and a weight scale; the host comprises a display, a camera and an operating console; the operating platform comprises an identity recognition device, a controller and a printer, the display, the camera, the identity recognition device and the printer are respectively and electrically connected with the controller, and the weighing scale is in communication connection with the controller;
the display is arranged at the upper part of the host machine, the camera is arranged at the middle part of the host machine, and the operating platform is arranged at the lower part of the host machine; the identity recognition module comprises a card reader and a scanner.
5. The physical data acquisition apparatus according to claim 4, characterized in that: the display is a touch display screen or a projector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910316421.8A CN110074788B (en) | 2019-04-18 | 2019-04-18 | Body data acquisition method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910316421.8A CN110074788B (en) | 2019-04-18 | 2019-04-18 | Body data acquisition method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110074788A CN110074788A (en) | 2019-08-02 |
| CN110074788B true CN110074788B (en) | 2020-03-17 |
Family
ID=67415721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910316421.8A Active CN110074788B (en) | 2019-04-18 | 2019-04-18 | Body data acquisition method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110074788B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110411308B (en) * | 2019-08-14 | 2021-08-24 | 浙江德尔达医疗科技有限公司 | Precision detection method for customized 3D printing model |
| CN110782482A (en) * | 2019-10-21 | 2020-02-11 | 深圳市网心科技有限公司 | Motion evaluation method and device, computer equipment and storage medium |
| CN111311732B (en) * | 2020-04-26 | 2023-06-20 | 中国人民解放军国防科技大学 | 3D human body grid acquisition method and device |
| CN112819881B (en) * | 2021-01-29 | 2023-10-31 | 福州靠谱云科技有限公司 | Human body measuring method |
| CN112733804B (en) * | 2021-01-29 | 2024-01-19 | 闽江学院 | Image pick-up device for measuring human body parameters |
| CN113112321A (en) * | 2021-03-10 | 2021-07-13 | 深兰科技（上海）有限公司 | Intelligent energy body method, device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002259474A (en) * | 2001-03-05 | 2002-09-13 | Oojisu Soken:Kk | Method and device for generating human body model, computer program and recording medium |
| CN105653742A (en) * | 2014-11-10 | 2016-06-08 | 江苏中佑石油机械科技有限责任公司 | Clothes model building method in three-dimension simulation fitting system |
| CN107545598A (en) * | 2017-07-31 | 2018-01-05 | 深圳市蒜泥科技有限公司 | A kind of human 3d model synthesis and body data acquisition methods |
| CN107705355A (en) * | 2017-09-08 | 2018-02-16 | 郭睿 | A kind of 3D human body modeling methods and device based on plurality of pictures |
| CN107924579A (en) * | 2015-08-14 | 2018-04-17 | 麦特尔有限公司 | The method for generating personalization 3D head models or 3D body models |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9613424B2 (en) * | 2013-09-23 | 2017-04-04 | Beihang University | Method of constructing 3D clothing model based on a single image |
-
2019
- 2019-04-18 CN CN201910316421.8A patent/CN110074788B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002259474A (en) * | 2001-03-05 | 2002-09-13 | Oojisu Soken:Kk | Method and device for generating human body model, computer program and recording medium |
| CN105653742A (en) * | 2014-11-10 | 2016-06-08 | 江苏中佑石油机械科技有限责任公司 | Clothes model building method in three-dimension simulation fitting system |
| CN107924579A (en) * | 2015-08-14 | 2018-04-17 | 麦特尔有限公司 | The method for generating personalization 3D head models or 3D body models |
| CN107545598A (en) * | 2017-07-31 | 2018-01-05 | 深圳市蒜泥科技有限公司 | A kind of human 3d model synthesis and body data acquisition methods |
| CN107705355A (en) * | 2017-09-08 | 2018-02-16 | 郭睿 | A kind of 3D human body modeling methods and device based on plurality of pictures |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110074788A (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110074788B (en) | Body data acquisition method and device based on machine learning | |
| CN107358648B (en) | Real-time full-automatic high quality three-dimensional facial reconstruction method based on individual facial image | |
| US11017547B2 (en) | Method and system for postural analysis and measuring anatomical dimensions from a digital image using machine learning | |
| US11010896B2 (en) | Methods and systems for generating 3D datasets to train deep learning networks for measurements estimation | |
| CN107194987B (en) | Method for predicting human body measurement data | |
| CN101322589B (en) | Non-contact type human body measuring method for clothing design | |
| US6490534B1 (en) | Camera measurement system | |
| US20150347833A1 (en) | Noncontact Biometrics with Small Footprint | |
| CN107041585A (en) | The measuring method of human dimension | |
| CN106981091B (en) | Human body three-dimensional modeling data processing method and device | |
| Tsoli et al. | Model-based anthropometry: Predicting measurements from 3D human scans in multiple poses | |
| CN107016721A (en) | The modeling method of human 3d model | |
| CN104794722A (en) | Dressed human body three-dimensional bare body model calculation method through single Kinect | |
| CN102439603A (en) | Simple techniques for three-dimensional modeling | |
| CN109919141A (en) | A kind of recognition methods again of the pedestrian based on skeleton pose | |
| US20220351378A1 (en) | Methods and systems for generating 3d datasets to train deep learning networks for measurements estimation | |
| CN111012353A (en) | Height detection method based on face key point recognition | |
| CN112102947B (en) | Apparatus and method for body posture assessment | |
| CN109655011B (en) | Method and system for measuring dimension of human body modeling | |
| CN107016697A (en) | A kind of height measurement method and device | |
| CN112016497A (en) | Single-view Taijiquan action analysis and assessment system based on artificial intelligence | |
| CN110477921B (en) | Height measurement method based on skeleton broken line Ridge regression | |
| Wang et al. | Facial feature extraction in an infrared image by proxy with a visible face image | |
| CN112116702A (en) | Human body measurement three-dimensional modeling method and system based on front and side profiles | |
| CN109740458B (en) | Method and system for measuring physical characteristics based on video processing |
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 | ||
| CP03 | Change of name, title or address |
Address after: 266000 802, Building 3, Publishing Building, No. 182, Haier Road, Laoshan District, Qingdao, Shandong Patentee after: Shandong Bihai Clothing Co.,Ltd. Address before: 266000 Intersection of Taiguang Road and Yangchun Road, Jiangshan Town, Laixi City, Qingdao, Shandong Province Patentee before: MENGDUO TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address |