CN107273846A - A kind of human somatotype parameter determination method and device - Google Patents
A kind of human somatotype parameter determination method and device Download PDFInfo
- Publication number
- CN107273846A CN107273846A CN201710439904.8A CN201710439904A CN107273846A CN 107273846 A CN107273846 A CN 107273846A CN 201710439904 A CN201710439904 A CN 201710439904A CN 107273846 A CN107273846 A CN 107273846A
- Authority
- CN
- China
- Prior art keywords
- human body
- human
- key point
- point
- body 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention provides a kind of human somatotype parameter determination method and device, and methods described includes：The first human body image and the second human body image of binocular camera shooting system photographs are obtained, wherein being respectively provided with object of reference in the first human body image and the second human body image；At least one human body key point is recognized from the first human body image and the second human body image respectively and the human region profile of human body key point is covered, and the key point and human region profile to identifying are matched；The depth being projected according to the marginal point on the matching result and human region profile of the marginal point on the depth of the matching result of key point and key point, human region profile, determines human body three-dimensional data；Body local spot size is determined according to human body three-dimensional data；Calculate the size of object of reference profile and the ratio of body local spot size；The full size information of proportion of utilization and the object of reference prestored calculates human somatotype parameter.
Description
Technical field
The present invention relates to human parameters fields of measurement, and in particular to a kind of human somatotype parameter determination method and device.
Background technology
Current internet+in the environment of, " individual customization " increasingly popularizes.In the demand of creation, shoes and hats
In, will require that customization customer is visited and measure without exception, with obtain special human somatotype parameter (such as waistline,
Shoulder breadth, brachium etc.) it is customized work.Although individual's customization meets the individual demand of people, measurement is visited in itself
It is troublesome, do not meet yet internet+definition.
Some are had at present the method for human somatotype parameter is determined by shooting image, but be required for user according to non-
Often then the Multi-angle human of the parameter to shoot such as harsh shooting angle, distance can just be modeled as picture by computer three-dimensional
Technology completes to set up the human 3d model for being photographed people, you can obtain human somatotype parameter.But user and non-professional
Camera work personnel, therefore user is difficult to grasp photographic parameter required in existing method, therefore the figure shot using user
As calculating to human somatotype parameter, due to uncertain, the obtained people of the parameters such as the shooting distance of image, angle, inclination
Body shape parameter is inaccurate, or even can not be calculated.
The content of the invention
Therefore, the technical problem to be solved in the present invention is the body for overcoming utilization image of the prior art to calculate
The inaccurate defect of shape parameter.
The present invention provides a kind of human somatotype parameter determination method, including：
The first human body image and the second human body image of binocular camera shooting system photographs are obtained, wherein the first body image
Object of reference is respectively provided with the second human body image；
At least one human body key point and covering are recognized from first human body image and the second human body image respectively
The human region profile of the human body key point, and key point and human region profile to identifying match；
The depth of key point is determined according to the matching result of predetermined binocular camera shooting system camera parameter and key point；
The edge on human region profile that will be identified respectively from first human body image and the second human body image
Spot projection in the depth of the key point to be matched to the marginal point on human region profile；
According to the matching result of the marginal point on the depth of the matching result of key point and key point, human region profile and
The depth that marginal point on human region profile is projected, determines human body three-dimensional data；
Body local spot size is determined according to the human body three-dimensional data；
The profile of the object of reference is recognized from first human body image and/or second human body image；
Calculate the size of the object of reference profile and the ratio of the body local spot size；
Human somatotype parameter is calculated using the ratio and the full size information of the object of reference prestored.
Preferably, the key point be the crown, sole, hand end points, waist both sides end points, chest both sides end points, it is described
Human region profile is head zone, leg area, arm region, lumbar region, chest area, the human body three-dimensional packet
Three-dimensional data, the edge three dimensional point cloud of human region profile of key point are included, the human somatotype parameter includes height, arm
Length, leg length, waistline and bust.
Preferably, before the first human body image and the second human body image of the acquisition binocular camera shooting system photographs, also
Including：Generate for pointing out user the prompt message shot using predetermined object of reference, the prompt message includes object of reference
Placement location and human body shoot posture.
Preferably, it is described to recognize that at least one human body is crucial from first human body image and the second human body image respectively
Point and the human region profile for covering the human body key point, including：
Each key point is recognized according to human body overall profile in first human body image and the second human body image respectively；
Covering institute is determined according to the position of the key point in first human body image and the second human body image respectively
State the human region profile of key point.
Preferably, the depth of the matching result and key point according to key point, the marginal point on human region profile
Matching result and human region profile on the depth that is projected of marginal point, determine human body three-dimensional data, including：
The three-dimensional data of key point is determined according to the depth of the matching result of the key point and key point；
People is set up according to the depth that the matching result and the marginal point of the marginal point of the human region profile are projected
Body region contoured three-dimensional cloud data, wherein the pass that the depth that the marginal point is projected is covered by the human region profile
The depth of key point.
Correspondingly, the present invention also provides a kind of human somatotype parameter determining device, including：
Acquiring unit, wherein the first human body image and the second human body image for obtaining binocular camera shooting system photographs, institute
State in the first human body image and the second human body image and be respectively provided with object of reference；
Human body recognition unit, for recognizing at least one from first human body image and the second human body image respectively
The human region profile of individual human body key point and the covering human body key point, and key point and human region to identifying
Profile is matched；
Depth determining unit, for being determined according to the matching result of predetermined binocular camera shooting system camera parameter and key point
The depth of key point；
Matching unit, for the human region that will be identified respectively from first human body image and the second human body image
Marginal point on profile projects in the depth of the key point to match the marginal point on human region profile；
Three-dimensional information determining unit, for matching result and the depth of key point, the human region profile according to key point
On marginal point matching result and human region profile on the depth that is projected of marginal point, determine human body three-dimensional data；
Picture size determining unit, for determining body local spot size according to the human body three-dimensional data；
Object of reference recognition unit, it is described for being recognized from first human body image and/or second human body image
The profile of object of reference；
Ratio computing unit, for calculating the size of the object of reference profile and the ratio of the body local spot size
Example；
Actual size computing unit, based on using the ratio and the full size information of the object of reference prestored
Calculate human somatotype parameter.
Preferably, the key point be the crown, sole, hand end points, waist both sides end points, chest both sides end points, it is described
Human region profile is head zone, leg area, arm region, lumbar region, chest area, the human body three-dimensional packet
Three-dimensional data, the edge three dimensional point cloud of human region profile of key point are included, the human somatotype parameter includes height, arm
Length, leg length, waistline and bust.
Preferably, in addition to：Tip element, for before acquiring unit work, generating for pointing out user to make
The prompt message shot with predetermined object of reference, the prompt message includes object of reference placement location and human body shoots posture.
Preferably, the human body recognition unit includes：
Key point recognition unit, for respectively in first human body image and the second human body image according to human body entirety
Each key point of outline identification；
Human region outline identification unit, for respectively in first human body image and the second human body image according to institute
The position for stating key point determines the human region profile of the covering key point.
Preferably, the three-dimensional information determining unit includes：
Key point three-dimensional information determining unit, the depth for the matching result according to the key point and key point is determined
The three-dimensional data of key point；
Region contour three-dimensional point cloud determining unit, for the marginal point according to the human region profile matching result and
The depth that the marginal point is projected sets up human region contoured three-dimensional cloud data, wherein the depth that the marginal point is projected
The depth of the key point covered by the human region profile.
The human somatotype parameter determination method and device provided according to the present invention, in the first-hand of binocular camera shooting system photographs
Key point and human region profile are recognized in portion's image and the second hand images, key point is matched according to body shape,
It is possible thereby to determine the depth of key point；Then the marginal point of human region profile is projected in the depth of key point,
Each marginal point of human region profile is matched in the depth plane of determination, and it is possible thereby to according to matching result and depth
Degree determines the three-dimensional data of partes corporis humani position.Human body contour outline and object of reference profile are recognized in body image afterwards, due to figure
The ratio of object of reference profile and human body contour outline as in is that full size can be calculated, object of reference is known and fixed
, it is possible thereby to which it is actual to calculate human body using the full size of the ratio and object of reference at object of reference profile and body local position
Shape parameter, the human somatotype parameter thus measured is more accurate.
Brief description of the drawings
, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical scheme of the prior art
The accompanying drawing used required in embodiment or description of the prior art is briefly described, it should be apparent that, in describing below
Accompanying drawing is some embodiments of the present invention, for those of ordinary skill in the art, before creative work is not paid
Put, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is the flow chart of human somatotype parameter determination method provided in an embodiment of the present invention；
Fig. 2 is the structure chart of human somatotype parameter determining device provided in an embodiment of the present invention.
Embodiment
Technical scheme is clearly and completely described below in conjunction with accompanying drawing, it is clear that described implementation
Example is a part of embodiment of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill
The every other embodiment that personnel are obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
As long as in addition, technical characteristic involved in invention described below different embodiments non-structure each other
It can just be combined with each other into conflict.
Embodiment 1
The embodiment of the present invention provides a kind of human somatotype parameter determination method, as shown in figure 1, including：
S1, obtains the first human body image and the second human body image of binocular camera shooting system photographs, wherein the first body
Object of reference is respectively provided with image and the second human body image.Wherein binocular camera shooting system is also referred to as binocular vision system, system tool
There are 2 imaging devices, based on principle of parallax and using 2 imaging devices from two width figures of different position acquisition testees
Picture, i.e. the first human body image and the second human body image.Binocular vision system has a variety of, utilizes any existing system photographs pair
It is all feasible that human body, which carries out shooting 2 images of acquisition,.
S2, recognizes at least one human body key point and covers from first human body image and the second human body image respectively
Cover the human region profile of the human body key point, and the key point and human region profile to identifying are matched.For example,
Target point is crown point, and target area is head.Above-mentioned 2 human body images are two dimensional image, and symbol is identified from two dimensional image
Closing the mode of the target of predetermined characteristic has a variety of, is all feasible using existing mode, thus identify that target point and target
The purpose in region is to be easily achieved.On matching, in stereoscopic vision, images match refers to any in three dimensions or one
Picture point or region of the individual region on the imaging surface of left and right cameras are mapped.
By taking the point of the crown as an example, by identifying processing, crown point aL and aR can be identified respectively in above-mentioned 2 image, by
There was only 1 head summit respectively in 2 images, therefore directly aL and aR can be mapped；Similarly, it will directly can also cover
The head zone on headkerchief summit is mapped.For sole point and leg, hand end points and arm, then it needs to be determined that left and right leg,
Right-hand man could mutually correspond to, and can specifically be recognized according to the characteristic of human leg's shape and arm shape and match target point
Or target area.
S3, the depth of key point is determined according to the matching result of predetermined binocular camera shooting system camera parameter and key point；
Specifically coordinate system can be set up using the midpoint of the binocular line of binocular vision system as origin, wherein, with binocular camera shooting system
The parallel plane of imaging plane is X/Y plane, and the direction vertical with this plane is Z-direction, and Z-direction is depth direction, target
The depth value of point is exactly the Z coordinate value under this coordinate system.After matching result is determined, then it can be existed based on same object point
Subpoint on the camera of left and right, by binocular camera parameter, recovers the wide-angle distortion of video camera, recycle focal length of camera,
The parameters such as left and right cameras interval, can calculate the depth information of this object point according to the geometry imaging relations of camera.
S4, the side on human region profile that will be identified respectively from first human body image and the second human body image
Edge spot projection in the depth of the key point to be matched to the marginal point on human region profile.Human region profile is
It is made up of many points, this step is only handled the marginal point in region, by taking arm and hand end points as an example, due in step
The depth of hand end points has been determined in rapid S3, thus a plane can be determined based on the depth, then by arm region
Marginal point all project in the depth plane, the arm region in above-mentioned 2 image is projected in the plane, the plane
On then have 2 arm regions, then point immediate to each pair in 2 arm regions is matched, and thus can be achieved to arm
All marginal points in arm region are matched.
S5, according to the matching knot of the marginal point on the depth of the matching result of key point and key point, human region profile
The depth that marginal point on fruit and human region profile is projected, determines human body three-dimensional data.Wherein described three-dimensional data includes
But it is not limited to the three-dimensional data of key point, edge three dimensional point cloud (all marginal points in a region of human region profile
Three-dimensional data the logical three dimensional point cloud for being referred to as the edges of regions of set).
S6, body local spot size is determined according to the human body three-dimensional data.Such as arm length in the picture, leg
Length, height, shoulder breadth in the picture etc..
S7, recognizes the profile of the object of reference from first human body image and/or second human body image.Utilize
Image processing meanses find object of reference profile from figure, in order to reduce identification difficulty, can point out user in solid color
Background under shot, here is omitted.
S8, calculates the size of the object of reference profile and the ratio of the body local spot size.Such as dimensional units
It is pixel, it is assumed that the object of reference profile in picture is longitudinally 60, is laterally 90, and shot image is shone for human body half body, human body contour outline
Height (longitudinal direction most strong point) be 450, a size in the width and length of object of reference profile can be chosen, with human body contour outline
Height compared to a ratio is obtained, such as ratio is 90/450=1/5.
S9, human somatotype parameter is calculated using the ratio and the full size information of the object of reference prestored.It is false
If the length in kind (transverse direction) of object of reference is 16cm, the actual height that calculating 16*5 can calculate the measured in image is
80cm, is that half body shines due to what is shot in the present embodiment, so this height value is upper half of human body length.
The human somatotype parameter determination method provided according to the present invention, in the first hand images of binocular camera shooting system photographs
With identification key point and human region profile in the second hand images, key point is matched according to body shape, thus may be used
To determine the depth of key point；Then the marginal point of human region profile is projected in the depth of key point, it is determined that
Each marginal point of human region profile is matched in depth plane, and it is possible thereby to determined according to matching result and depth
The three-dimensional data of partes corporis humani position.Human body contour outline and object of reference profile are recognized in body image afterwards, due in image
The ratio of object of reference profile and human body contour outline is that full size can be calculated, object of reference is known and fixed, thus
The actual build of human body can be calculated using the full size of the ratio and object of reference at object of reference profile and body local position to join
Number, the human somatotype parameter thus measured is more accurate.
As one preferred embodiment, the key point is the crown, sole, hand end points, waist both sides end points, chest
Portion both sides end points, the human region profile is head zone, leg area, arm region, lumbar region, chest area, institute
State three-dimensional data of the human body three-dimensional data including key point, the edge three dimensional point cloud of human region profile, the body
Shape parameter includes height, brachium, leg length, waistline and bust.
As one preferred embodiment, in the first human body image and second of the acquisition binocular camera shooting system photographs
Before human body image, in addition to：Generate for pointing out user the prompt message shot using predetermined object of reference, the prompting
Information includes object of reference placement location and human body shoots posture.
As one preferred embodiment, it is described to be recognized respectively from first human body image and the second human body image
The human region profile of at least one human body key point and the covering human body key point, including：
Each key point is recognized according to human body overall profile in first human body image and the second human body image respectively；
Covering institute is determined according to the position of the key point in first human body image and the second human body image respectively
State the human region profile of key point.
As one preferred embodiment, the matching result and depth, the human body area of key point according to key point
The depth that marginal point on the matching result and human region profile of marginal point on the profile of domain is projected, determines human body three-dimensional number
According to, including：
The three-dimensional data of key point is determined according to the depth of the matching result of the key point and key point；
People is set up according to the depth that the matching result and the marginal point of the marginal point of the human region profile are projected
Body region contoured three-dimensional cloud data, wherein the pass that the depth that the marginal point is projected is covered by the human region profile
The depth of key point.
Embodiment 2
The embodiment of the present invention provides a kind of human somatotype parameter determining device, as shown in Fig. 2 including：
Acquiring unit 21, the first human body image and the second human body image for obtaining binocular camera shooting system photographs, wherein
Object of reference is respectively provided with first human body image and the second human body image；
Human body recognition unit 22, for being recognized at least from first human body image and the second human body image respectively
The human region profile of one human body key point and the covering human body key point, and the key point to identifying and human body area
Domain profile is matched；
Depth determining unit 23, it is true for the matching result according to predetermined binocular camera shooting system camera parameter and key point
Determine the depth of key point；
Matching unit 24, for the human body area that will be identified respectively from first human body image and the second human body image
Marginal point on the profile of domain projects in the depth of the key point to match the marginal point on human region profile；
Three-dimensional information determining unit 25, for matching result and the depth of key point, the human region wheel according to key point
The depth that marginal point on the matching result and human region profile of marginal point on exterior feature is projected, determines human body three-dimensional data；
Picture size determining unit 26, for determining body local spot size according to the human body three-dimensional data；
Object of reference recognition unit 27, for recognizing institute from first human body image and/or second human body image
State the profile of object of reference；
Ratio computing unit 28, for calculating the size of the object of reference profile and the ratio of the body local spot size
Example；
Actual size computing unit 29, for the full size information using the ratio and the object of reference prestored
Calculate human somatotype parameter.
The human somatotype parameter determining device provided according to the present invention, in the first hand images of binocular camera shooting system photographs
With identification key point and human region profile in the second hand images, key point is matched according to body shape, thus may be used
To determine the depth of key point；Then the marginal point of human region profile is projected in the depth of key point, it is determined that
Each marginal point of human region profile is matched in depth plane, and it is possible thereby to determined according to matching result and depth
The three-dimensional data of partes corporis humani position.Human body contour outline and object of reference profile are recognized in body image afterwards, due in image
The ratio of object of reference profile and human body contour outline is that full size can be calculated, object of reference is known and fixed, thus
The actual build of human body can be calculated using the full size of the ratio and object of reference at object of reference profile and body local position to join
Number, the human somatotype parameter thus measured is more accurate.
As one preferred embodiment, the key point is the crown, sole, hand end points, waist both sides end points, chest
Portion both sides end points, the human region profile is head zone, leg area, arm region, lumbar region, chest area, institute
State three-dimensional data of the human body three-dimensional data including key point, the edge three dimensional point cloud of human region profile, the body
Shape parameter includes height, brachium, leg length, waistline and bust.
As one preferred embodiment, in addition to：Tip element, it is raw for before acquiring unit work
Into for pointing out user the prompt message shot using predetermined object of reference, the prompt message includes object of reference placement location
Posture is shot with human body.
As one preferred embodiment, the human body recognition unit includes：
Key point recognition unit, for respectively in first human body image and the second human body image according to human body entirety
Each key point of outline identification；
Human region outline identification unit, for respectively in first human body image and the second human body image according to institute
The position for stating key point determines the human region profile of the covering key point.
As one preferred embodiment, the three-dimensional information determining unit includes：
Key point three-dimensional information determining unit, the depth for the matching result according to the key point and key point is determined
The three-dimensional data of key point；
Region contour three-dimensional point cloud determining unit, for the marginal point according to the human region profile matching result and
The depth that the marginal point is projected sets up human region contoured three-dimensional cloud data, wherein the depth that the marginal point is projected
The depth of the key point covered by the human region profile.
Obviously, above-described embodiment is only intended to clearly illustrate example, and the not restriction to embodiment.It is right
For those of ordinary skill in the art, can also make on the basis of the above description it is other it is various forms of change or
Change.There is no necessity and possibility to exhaust all the enbodiments.And the obvious change thus extended out or
Among changing still in the protection domain of the invention.
Claims (10)
1. a kind of human somatotype parameter determination method, it is characterised in that including：
The first human body image and the second human body image of binocular camera shooting system photographs are obtained, wherein the first body image and the
Object of reference is respectively provided with two human body images；
Recognize that at least one human body key point and covering are described from first human body image and the second human body image respectively
The human region profile of human body key point, and key point and human region profile to identifying match；
The depth of key point is determined according to the matching result of predetermined binocular camera shooting system camera parameter and key point；
Marginal point on the human region profile that will be identified respectively from first human body image and the second human body image is thrown
Shadow in the depth of the key point to be matched to the marginal point on human region profile；
According to the matching result and human body of the marginal point on the depth of the matching result of key point and key point, human region profile
The depth that marginal point on region contour is projected, determines human body three-dimensional data；
Body local spot size is determined according to the human body three-dimensional data；
The profile of the object of reference is recognized from first human body image and/or second human body image；
Calculate the size of the object of reference profile and the ratio of the body local spot size；
Human somatotype parameter is calculated using the ratio and the full size information of the object of reference prestored.
2. according to the method described in claim 1, it is characterised in that the key point is the crown, sole, hand end points, waist
Both sides end points, chest both sides end points, the human region profile be head zone, leg area, arm region, lumbar region,
Chest area, the human body three-dimensional data include the three-dimensional data of key point, the edge three dimensional point cloud of human region profile,
The human somatotype parameter includes height, brachium, leg length, waistline and bust.
3. according to the method described in claim 1, it is characterised in that in the first human body of the acquisition binocular camera shooting system photographs
Before image and the second human body image, in addition to：Generate and believe for the prompting for pointing out user to be shot using predetermined object of reference
Breath, the prompt message includes object of reference placement location and human body shoots posture.
4. the method according to any one of claim 1-3, it is characterised in that described respectively from first human body image
With the human region profile that at least one human body key point and the covering human body key point are recognized in the second human body image, bag
Include：
Each key point is recognized according to human body overall profile in first human body image and the second human body image respectively；
Determine that covering is described according to the position of the key point in first human body image and the second human body image respectively to close
The human region profile of key point.
5. the method according to any one of claim 1-3, it is characterised in that the matching result according to key point and
What the marginal point in the depth of key point, the matching result and human region profile of the marginal point on human region profile was projected
Depth, determines human body three-dimensional data, including：
The three-dimensional data of key point is determined according to the depth of the matching result of the key point and key point；
Human body area is set up according to the depth that the matching result and the marginal point of the marginal point of the human region profile are projected
Domain contoured three-dimensional cloud data, wherein the key point that the depth that the marginal point is projected is covered by the human region profile
Depth.
6. a kind of human somatotype parameter determining device, it is characterised in that including：
Acquiring unit, the first human body image and the second human body image for obtaining binocular camera shooting system photographs, wherein described
Object of reference is respectively provided with one human body image and the second human body image；
Human body recognition unit, for recognizing at least one people from first human body image and the second human body image respectively
The human region profile of body key point and the covering human body key point, and key point and human region profile to identifying
Matched；
Depth determining unit, it is crucial for being determined according to the matching result of predetermined binocular camera shooting system camera parameter and key point
The depth of point；
Matching unit, for the human region profile that will be identified respectively from first human body image and the second human body image
On marginal point project in the depth of the key point to match the marginal point on human region profile；
Three-dimensional information determining unit, on the matching result and the depth of key point according to key point, human region profile
The depth that marginal point on the matching result and human region profile of marginal point is projected, determines human body three-dimensional data；
Picture size determining unit, for determining body local spot size according to the human body three-dimensional data；
Object of reference recognition unit, for recognizing the reference from first human body image and/or second human body image
The profile of thing；
Ratio computing unit, for calculating the size of the object of reference profile and the ratio of the body local spot size；
Actual size computing unit, for calculating people using the ratio and the full size information of the object of reference prestored
Body shape parameter.
7. device according to claim 6, it is characterised in that the key point is the crown, sole, hand end points, waist
Both sides end points, chest both sides end points, the human region profile be head zone, leg area, arm region, lumbar region,
Chest area, the human body three-dimensional data include the three-dimensional data of key point, the edge three dimensional point cloud of human region profile,
The human somatotype parameter includes height, brachium, leg length, waistline and bust.
8. device according to claim 6, it is characterised in that also include：Tip element, in the acquiring unit work
Before work, generate for pointing out user the prompt message shot using predetermined object of reference, the prompt message includes reference
Thing placement location and human body shoot posture.
9. the device according to any one of claim 6-8, it is characterised in that the human body recognition unit includes：
Key point recognition unit, for respectively in first human body image and the second human body image according to human body overall profile
Recognize each key point；
Human region outline identification unit, for being closed respectively in first human body image and the second human body image according to described
The position of key point determines the human region profile of the covering key point.
10. the device according to any one of claim 6-8, it is characterised in that the three-dimensional information determining unit includes：
Key point three-dimensional information determining unit, the depth for the matching result according to the key point and key point determines crucial
The three-dimensional data of point；
Region contour three-dimensional point cloud determining unit, for the matching result of the marginal point according to the human region profile and described
The depth that marginal point is projected sets up human region contoured three-dimensional cloud data, wherein the depth that the marginal point is projected is institute
State the depth for the key point that human region profile is covered.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710439904.8A CN107273846B (en) | 2017-06-12 | 2017-06-12 | Human body shape parameter determination method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710439904.8A CN107273846B (en) | 2017-06-12 | 2017-06-12 | Human body shape parameter determination method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107273846A true CN107273846A (en) | 2017-10-20 |
| CN107273846B CN107273846B (en) | 2020-08-07 |
Family
ID=60066170
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710439904.8A Active CN107273846B (en) | 2017-06-12 | 2017-06-12 | Human body shape parameter determination method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107273846B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038469A (en) * | 2017-12-27 | 2018-05-15 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting human body |
| CN108765367A (en) * | 2018-04-16 | 2018-11-06 | 武汉纺织大学 | A kind of dressing human dimension determines method and system |
| CN108830200A (en) * | 2018-05-31 | 2018-11-16 | 北京市商汤科技开发有限公司 | A kind of image processing method, device and computer storage medium |
| CN108903237A (en) * | 2018-07-23 | 2018-11-30 | 广州维纳斯家居股份有限公司 | Intelligent elevated table height adjusting method, device, intelligent elevated table and storage medium |
| CN109508708A (en) * | 2019-01-15 | 2019-03-22 | 广州唯品会研究院有限公司 | Measurement method, device and the computer readable storage medium of body data |
| CN109766755A (en) * | 2018-12-06 | 2019-05-17 | 深圳市天彦通信股份有限公司 | Face identification method and Related product |
| CN109840500A (en) * | 2019-01-31 | 2019-06-04 | 深圳市商汤科技有限公司 | A kind of 3 D human body posture information detection method and device |
| CN110037705A (en) * | 2019-04-25 | 2019-07-23 | 北京新睿搏创科技有限公司 | A kind of method and system measuring human dimension |
| CN110123331A (en) * | 2019-04-17 | 2019-08-16 | 平安科技（深圳）有限公司 | Human body body and constitution collecting method, device and storage medium |
| CN110443188A (en) * | 2019-07-31 | 2019-11-12 | 上海悟应物联网技术有限公司 | The human body image identifying system and method for dress designing and manufacture under big data environment |
| CN110728795A (en) * | 2019-10-31 | 2020-01-24 | 赵飞腾 | Intelligent coffee dosage adjusting system |
| CN110852943A (en) * | 2018-08-21 | 2020-02-28 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, image processing device, and storage medium |
| CN111340042A (en) * | 2018-12-19 | 2020-06-26 | 中国移动通信集团四川有限公司 | Object contour recognition method, device, equipment and storage medium |
| CN112329723A (en) * | 2020-11-27 | 2021-02-05 | 北京邮电大学 | Binocular camera-based multi-person human body 3D skeleton key point positioning method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337094A (en) * | 2013-06-14 | 2013-10-02 | 西安工业大学 | Method for realizing three-dimensional reconstruction of movement by using binocular camera |
| CN104966284A (en) * | 2015-05-29 | 2015-10-07 | 北京旷视科技有限公司 | Method and equipment for acquiring object dimension information based on depth data |
| US20160110595A1 (en) * | 2014-10-17 | 2016-04-21 | Qiaosong Wang | Fast 3d model fitting and anthropometrics using synthetic data |
| CN106580329A (en) * | 2016-12-28 | 2017-04-26 | 天津众阳科技有限公司 | Height measurement system and method based on binocular stereovision technology |
| CN106666903A (en) * | 2015-11-09 | 2017-05-17 | 恒基伟业投资发展集团有限公司 | Method and device utilizing images to measure human body shape parameter |
-
2017
- 2017-06-12 CN CN201710439904.8A patent/CN107273846B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337094A (en) * | 2013-06-14 | 2013-10-02 | 西安工业大学 | Method for realizing three-dimensional reconstruction of movement by using binocular camera |
| US20160110595A1 (en) * | 2014-10-17 | 2016-04-21 | Qiaosong Wang | Fast 3d model fitting and anthropometrics using synthetic data |
| CN104966284A (en) * | 2015-05-29 | 2015-10-07 | 北京旷视科技有限公司 | Method and equipment for acquiring object dimension information based on depth data |
| CN106666903A (en) * | 2015-11-09 | 2017-05-17 | 恒基伟业投资发展集团有限公司 | Method and device utilizing images to measure human body shape parameter |
| CN106580329A (en) * | 2016-12-28 | 2017-04-26 | 天津众阳科技有限公司 | Height measurement system and method based on binocular stereovision technology |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108038469A (en) * | 2017-12-27 | 2018-05-15 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting human body |
| CN108038469B (en) * | 2017-12-27 | 2019-10-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for detecting human body |
| CN108765367A (en) * | 2018-04-16 | 2018-11-06 | 武汉纺织大学 | A kind of dressing human dimension determines method and system |
| CN108830200A (en) * | 2018-05-31 | 2018-11-16 | 北京市商汤科技开发有限公司 | A kind of image processing method, device and computer storage medium |
| CN108903237A (en) * | 2018-07-23 | 2018-11-30 | 广州维纳斯家居股份有限公司 | Intelligent elevated table height adjusting method, device, intelligent elevated table and storage medium |
| CN110852943A (en) * | 2018-08-21 | 2020-02-28 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, image processing device, and storage medium |
| CN109766755B (en) * | 2018-12-06 | 2020-11-27 | 深圳市天彦通信股份有限公司 | Face recognition method and related product |
| CN109766755A (en) * | 2018-12-06 | 2019-05-17 | 深圳市天彦通信股份有限公司 | Face identification method and Related product |
| CN111340042A (en) * | 2018-12-19 | 2020-06-26 | 中国移动通信集团四川有限公司 | Object contour recognition method, device, equipment and storage medium |
| CN111340042B (en) * | 2018-12-19 | 2023-09-22 | 中国移动通信集团四川有限公司 | Object contour recognition method, device, equipment and storage medium |
| CN109508708A (en) * | 2019-01-15 | 2019-03-22 | 广州唯品会研究院有限公司 | Measurement method, device and the computer readable storage medium of body data |
| CN109840500A (en) * | 2019-01-31 | 2019-06-04 | 深圳市商汤科技有限公司 | A kind of 3 D human body posture information detection method and device |
| CN109840500B (en) * | 2019-01-31 | 2021-07-02 | 深圳市商汤科技有限公司 | Three-dimensional human body posture information detection method and device |
| JP2021527877A (en) * | 2019-01-31 | 2021-10-14 | 深▲せん▼市商▲湯▼科技有限公司Ｓｈｅｎｚｈｅｎ Ｓｅｎｓｅｔｉｍｅ Ｔｅｃｈｎｏｌｏｇｙ Ｃｏ．， Ｌｔｄ． | 3D human body posture information detection method and devices, electronic devices, storage media |
| CN110123331A (en) * | 2019-04-17 | 2019-08-16 | 平安科技（深圳）有限公司 | Human body body and constitution collecting method, device and storage medium |
| CN110037705A (en) * | 2019-04-25 | 2019-07-23 | 北京新睿搏创科技有限公司 | A kind of method and system measuring human dimension |
| CN110443188A (en) * | 2019-07-31 | 2019-11-12 | 上海悟应物联网技术有限公司 | The human body image identifying system and method for dress designing and manufacture under big data environment |
| WO2021018269A1 (en) * | 2019-07-31 | 2021-02-04 | 上海悟应物联网技术有限公司 | Human body image recognition system and method for clothing design and manufacturing in big data environment |
| CN110728795A (en) * | 2019-10-31 | 2020-01-24 | 赵飞腾 | Intelligent coffee dosage adjusting system |
| CN112329723A (en) * | 2020-11-27 | 2021-02-05 | 北京邮电大学 | Binocular camera-based multi-person human body 3D skeleton key point positioning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107273846B (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107273846A (en) | A kind of human somatotype parameter determination method and device | |
| CN106168853B (en) | A kind of free space wear-type gaze tracking system | |
| US11576645B2 (en) | Systems and methods for scanning a patient in an imaging system | |
| CN109816703B (en) | Point cloud registration method based on camera calibration and ICP algorithm | |
| US11576578B2 (en) | Systems and methods for scanning a patient in an imaging system | |
| CN104596502B (en) | Object posture measuring method based on CAD model and monocular vision | |
| US20150347833A1 (en) | Noncontact Biometrics with Small Footprint | |
| EP2870428B1 (en) | System and method for 3d measurement of the surface geometry of an object | |
| CN107169405A (en) | Method and device based on binocular camera vivo identification | |
| CN109035309A (en) | Pose method for registering between binocular camera and laser radar based on stereoscopic vision | |
| CN105740778B (en) | Improved three-dimensional human face in-vivo detection method and device | |
| JP5035467B2 (en) | Three-dimensional authentication method, three-dimensional authentication device, and three-dimensional authentication program | |
| CN105258680B (en) | A kind of object pose measuring method and device | |
| CN109357633A (en) | 3-D scanning method, apparatus, storage medium and processor | |
| CN110199296A (en) | Face identification method, processing chip and electronic equipment | |
| CN105740781A (en) | Three-dimensional human face in-vivo detection method and device | |
| CN108324247B (en) | Method and system for evaluating skin wrinkles at specified positions | |
| CN107374638A (en) | A kind of height measuring system and method based on binocular vision module | |
| CN105335699B (en) | Read-write scene is read and write intelligent identification and the application thereof of element three-dimensional coordinate | |
| BR112016030027B1 (en) | METHOD AND TERMINAL FOR OBTAINING SIGNAL DATA FROM A TARGET OBJECT | |
| KR101004629B1 (en) | Non-intrusive 3D face data acquisition system and method thereof | |
| Wang et al. | Facial feature extraction in an infrared image by proxy with a visible face image | |
| KR101557492B1 (en) | Apparatus and Method for generating user's three dimensional body model based on depth information | |
| CN110909571B (en) | High-precision face recognition space positioning method | |
| CN108197549A (en) | Face identification method and terminal based on 3D imagings |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Guoqiang Inventor after: Chen Juanfen Inventor before: Liu Jingjing Inventor before: Chen Guoqiang Inventor before: Chen Juanfen |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |