CN109409236A - Three-dimensional static gesture identification method and device - Google Patents
Three-dimensional static gesture identification method and device Download PDFInfo
- Publication number
- CN109409236A CN109409236A CN201811135368.3A CN201811135368A CN109409236A CN 109409236 A CN109409236 A CN 109409236A CN 201811135368 A CN201811135368 A CN 201811135368A CN 109409236 A CN109409236 A CN 109409236A
- Authority
- CN
- China
- Prior art keywords
- gesture
- dimension
- model
- finger joint
- 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
- 230000003068 static effect Effects 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 25
- 210000001145 finger joint Anatomy 0.000 claims description 42
- 238000004891 communication Methods 0.000 claims description 6
- 230000001360 synchronised effect Effects 0.000 claims description 4
- 238000013136 deep learning model Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 2
- 230000003993 interaction Effects 0.000 description 11
- 210000003811 finger Anatomy 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000005057 finger movement Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 210000003414 extremity Anatomy 0.000 description 1
- 210000004247 hand Anatomy 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- G06V40/107—Static hand or arm
- G06V40/113—Recognition of static hand signs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/017—Gesture based interaction, e.g. based on a set of recognized hand gestures
-
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- User Interface Of Digital Computer (AREA)
- Image Analysis (AREA)
Abstract
本发明提供了一种三维静态手势识别方法和装置，三维静态手势识别方法包括：获取手势正面图像；获取手势侧面图像；根据手势正面图像和手势侧面图像生成三维手势模型；对三维手势模型进行降维处理；根据降维后的三维手势模型进行手势识别。本发明不仅能够提高手势识别的精度和成功率，还能够降低手势识别的运算时间和功耗。
The invention provides a three-dimensional static gesture recognition method and device. The three-dimensional static gesture recognition method includes: acquiring a frontal image of the gesture; acquiring a side image of the gesture; generating a three-dimensional gesture model according to the frontal image of the gesture and the side image of the gesture; reducing the three-dimensional gesture model Dimensional processing; gesture recognition is performed according to the dimensionality-reduced 3D gesture model. The present invention can not only improve the accuracy and success rate of gesture recognition, but also reduce the operation time and power consumption of gesture recognition.
Description
Technical field
The present invention relates to human-computer interaction technique fields, and in particular to a kind of three-dimensional static gesture identification method and a kind of three-dimensional
Static gesture identification device.
Background technique
With the extensive use of computer, human-computer interaction has become the pith in people's daily life.Man-machine friendship
Mutual final goal is to realize that people exchanges naturally with machine, and wherein limbs Shaping language such as gesture, body posture and expression is people
Common exchange way, and gesture has the characteristics that convenient, lively, image and intuitive in human-computer interaction, therefore gesture is known
It is other to study the needs for having complied with development in science and technology.Currently, gesture identification mainly has two-dimentional gesture identification and three-dimension gesture identification etc.
Direction.With being gradually improved for Gesture Recognition, which is gradually used by various industries, greatly enhances people
The convenience and practicability of machine interaction.
But since manpower itself has the characteristics that diversity, ambiguity, variability and spatio-temporal difference, background environment
Complicated and changeable, unpredictable, gesture freedom degree is high and flexible in addition, chooses so that the gesture interaction of view-based access control model is rich in as one
The research topic of the multi-crossed disciplines of war property.The research of the gesture interaction technology and methods of view-based access control model, not only artificial intelligence,
There is important theory significance in the subjects such as pattern-recognition, machine learning, also have in intelligentized study, work and life
There is very extensive application.The gesture interaction technology of view-based access control model is to realize the indispensable Xiang Guanjian of human-computer interaction of new generation
Technology.And how to improve the precision and success rate of gesture identification, and reduce gesture identification operation time and power consumption become need
It solves the problems, such as.
Summary of the invention
The present invention be solve how to improve the precision and success rate of gesture identification, and reduce gesture identification operation time and
The technical issues of power consumption, provides a kind of three-dimensional static gesture identification method and device.
The technical solution adopted by the invention is as follows:
A kind of three-dimensional static gesture identification method, comprising: obtain gesture direct picture；Obtain gesture side image；According to
The gesture direct picture and the gesture side image generate three-dimension gesture model；Dimensionality reduction is carried out to the three-dimension gesture model
Processing；Gesture identification is carried out according to the three-dimension gesture model after dimensionality reduction.
The gesture direct picture is acquired by the first image acquisition units of corresponding target hand front setting, and is passed through
The second image acquisition units moved along the guide rail of corresponding target hand side setting are from multiple and different multiple hands of angle acquisition
Gesture side image.
The guide rail is arc-shaped guide rail, and the plane where the arc-shaped guide rail is perpendicular to where the palm of the target hand
Plane.
The first image acquisition unit acquires the primary gesture direct picture, and second image acquisition units are along institute
Arc-shaped guide rail is stated to move to the multiple gesture side images of second end acquisition from first end or move to first end acquisition from second end
Multiple gesture side images complete the synchronous acquisition of gesture direct picture and gesture side image.
Three-dimension gesture model is generated according to the gesture direct picture and the gesture side image, is specifically included: reference
The constraint relationship of three-dimensional hand skeleton model obtains three-dimension gesture model finger joint parameter according to the gesture direct picture；It calculates
The area of hand in each gesture side image, and the smallest gesture side image of area for choosing hand is as selected hand
Gesture side image；In conjunction with three-dimension gesture model finger joint parameter, three-dimension gesture model is obtained according to the selected gesture side image
Hold the angle parameter of finger joint and the plane of delineation；Refer in conjunction with the three-dimension gesture model finger joint parameter and the three-dimension gesture model end
The angle parameter of section and the plane of delineation generates the three-dimension gesture model.
The three-dimension gesture model finger joint parameter includes the length and root finger joint, middle finger joint, end finger joint of each finger finger joint
With the angle of the plane of delineation.
Dimension-reduction treatment is carried out to the three-dimension gesture model, is specifically included: three-dimension gesture model parameter vector is tieed up from 27
It is down to 12 dimensions.
Gesture identification is carried out according to the three-dimension gesture model after dimensionality reduction, is specifically included: with the three-dimension gesture model after dimensionality reduction
For input, three-dimension gesture model feature vector is obtained with the gesture identification deep learning model pre-established；To the three-dimensional
Gesture model feature vector is quantified, and obtains single gesture discrete features vector, and according to the single gesture discrete features
Vector is trained, and obtains static gesture classifier parameters；Static gesture classifier after training is used for the single hand
Gesture discrete features vector carries out classification judgement, exports gesture identification result.
A kind of three-dimensional static gesture identifying device, comprising: the first image acquisition units, the first image acquisition unit are used
In acquisition gesture direct picture；Second image acquisition units, second image acquisition units are for acquiring gesture side image；
Terminal, the terminal pass through with the first image acquisition unit and second image acquisition units respectively
Communication interface is connected, and the terminal is used to be generated according to the gesture direct picture and the gesture side image three-dimensional
Gesture model, and dimension-reduction treatment is carried out to the three-dimension gesture model, and hand is carried out according to the three-dimension gesture model after dimensionality reduction
Gesture identification.
The communication interface is USB (Universal Serial Bus, universal serial bus) data transmission interface or API
(Application Programming Interface, application programming interface) interface.
Beneficial effects of the present invention:
The present invention generates three-dimension gesture model by obtaining gesture direct picture and gesture side image, then to three
It ties up gesture model and carries out dimension-reduction treatment, and gesture identification is carried out according to the three-dimension gesture model after dimensionality reduction, hand can not only be improved
The precision and success rate of gesture identification, additionally it is possible to reduce operation time and the power consumption of gesture identification.
Detailed description of the invention
Fig. 1 is the flow chart of the three-dimensional static gesture identification method of the embodiment of the present invention；
Fig. 2 is the setting position view of the image acquisition units of one embodiment of the invention；
Fig. 3 is the schematic diagram that gesture side image is acquired with interval angles of one embodiment of the invention；
Fig. 4 is the schematic diagram that the image acquisition units of one embodiment of the invention move along the rail；
Fig. 5 is the flow chart of the three-dimensional static gesture identification method of a specific embodiment of the invention；
Fig. 6 is the block diagram of the three-dimensional static gesture identifying device of the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in Figure 1, the three-dimensional static gesture identification method of the embodiment of the present invention the following steps are included:
S1 obtains gesture direct picture.
In one embodiment of the invention, as shown in Fig. 2, the first image of corresponding target hand front setting can be passed through
Acquisition unit 1 acquires gesture direct picture.
S2 obtains gesture side image.
In one embodiment of the invention, as shown in Fig. 2, the guide rail 3 along corresponding target hand side setting can be passed through
Second image acquisition units 2 of movement are from multiple and different multiple gesture side images of angle acquisition.Wherein, guide rail 3 is arc
Guide rail, the plane where arc-shaped guide rail is perpendicular to the plane where the palm of target hand.It should be noted that target in Fig. 2
Positional relationship between hand and the first image acquisition units 1, guide rail 3 is only as signal, in a particular embodiment with the first figure
As the plane where the palm of 1 face target hand of acquisition unit one side, guide rail 3 is perpendicular to flat where the palm of target hand
Subject to face.
As shown in figure 3, when the second image acquisition units 2 are moved along guide rail 3, it can be according to preset angle acquisition hand
Gesture side image.N value is the angle number of acquisition, and N is integer and N > 1, it should be understood that in the certain situation of arc-shaped guide rail angle
Under, N value is bigger, and interval angles are smaller, and collected gesture side image is more；N value is smaller, and interval angles are bigger, collects
Gesture side image it is fewer.M value is any acquisition angles number, and M is integer and 1 < M < N.
It in one embodiment of the invention, can a synchronous acquisition gesture direct picture when carrying out gesture identification every time
With gesture side image.Specifically, the first image acquisition units acquire a gesture direct picture, the second image acquisition units edge
Arc-shaped guide rail moves to that second end acquires multiple gesture side images or to move to first end acquisition from second end more from first end
A gesture side image completes the synchronous acquisition of gesture direct picture and gesture side image.Second image acquisition units are completed
After the acquisition of multi-angle gesture side image, stop movement, when carrying out that images of gestures acquires next time, along guide rail to negative side
To movement.As shown in figure 4, the second image acquisition units 2 first time images of gestures moves clockwise when acquiring along guide rail 3,
Second end is moved to from first end, is moved for the second time along 3 counter clockwise direction of guide rail, moves to first end, third from second end
It is secondary to move clockwise along guide rail 3, and so on.
S3 generates three-dimension gesture model according to gesture direct picture and gesture side image.
S4 carries out dimension-reduction treatment to three-dimension gesture model.
Specifically, step S3 includes: the constraint relationship with reference to three-dimensional hand skeleton model, is obtained according to gesture direct picture
Three-dimension gesture model finger joint parameter；The area of hand in each gesture side image is calculated, and the area for choosing hand is the smallest
Gesture side image is as selected gesture side image；In conjunction with three-dimension gesture model finger joint parameter, according to selected gesture side view
As obtaining the angle parameter of three-dimension gesture model end finger joint and the plane of delineation；In conjunction with three-dimension gesture model finger joint parameter and three-dimensional hand
Potential model end finger joint and the angle parameter of the plane of delineation generate three-dimension gesture model.
Wherein, the smallest gesture side image of the area of hand, as most accurate gesture side image are more by acquiring
A gesture side image simultaneously determines most accurate gesture side image, can be improved the accuracy of gesture side image acquisition, from
And improve subsequent gesture identification precision.
In one embodiment of the invention, three-dimension gesture model finger joint parameter includes the length and root of each finger finger joint
Finger joint, middle finger joint, the angle for holding finger joint and the plane of delineation.
Further, hand Segmentation can be carried out to gesture direct picture and gesture side image, obtains normalized hand
Image.
The length and width of each finger finger joint can be obtained according to the collected gesture direct picture of the first image acquisition units
Degree, i.e., each finger are indicated that an angle indicates the angle of root finger joint and the plane of delineation by 2 angles, another angle indicates
The angle of middle finger joint and the plane of delineation, according to the linear restriction between root finger joint, middle finger joint angle and middle finger joint, end finger joint angle
Relation derivation obtains the angle of end finger joint and the plane of delineation, and manpower skeleton pattern is 27 freedom degrees.In addition each finger movement angle
The dynamic constrained between the static constraint and finger movement finger joint of range is spent, three-dimension gesture model parameter vector is down to from 27 dimensions
12 dimensions.Wherein, thumb freedom degree is reduced to 2 by 5, and middle finger freedom degree is reduced to 1 by 4, other three finger freedom degrees are reduced to 2 by 4,
The rotation angle of manpower indicates 3 freedom degrees, so, finally obtained three-dimension gesture model parameter vector is 12 dimensions.
S5 carries out gesture identification according to the three-dimension gesture model after dimensionality reduction.
It specifically, can be input with the three-dimension gesture model after dimensionality reduction, with the gesture identification deep learning pre-established
Model obtains three-dimension gesture model feature vector, wherein gesture identification deep learning model is the three-dimension gesture model of 12 dimensions.Again
Three-dimension gesture model feature vector is quantified, obtains single gesture discrete features vector, and according to the discrete spy of single gesture
Sign vector is trained, and obtains static gesture classifier parameters.Then, the static gesture classifier after training is used for single
Gesture discrete features vector carries out classification judgement, exports gesture identification result.
In one embodiment of the invention, above-mentioned step S3 to S5 can be executed by terminal.
In one particular embodiment of the present invention, as shown in figure 5, three-dimensional static gesture identification method includes following step
It is rapid:
S101, the first image acquisition units acquire gesture direct picture.
S102, the second image acquisition units do one-way movement along guide rail and case set angle acquires the hand of multiple and different angles
Gesture direct picture.
S103, terminal receive the first image acquisition units and the second image acquisition units acquired image.
S104 obtains three-dimension gesture model according to gesture direct picture with reference to the constraint relationship of three-dimensional hand skeleton model
Finger joint parameter.
S105 chooses the smallest gesture side image of area of hand as selected gesture side image.
S106 obtains three-dimension gesture model end according to selected gesture side image in conjunction with three-dimension gesture model finger joint parameter
The angle parameter of finger joint and the plane of delineation.
S107, in conjunction with the angle parameter of three-dimension gesture model finger joint parameter and three-dimension gesture model end finger joint and the plane of delineation
Generate three-dimension gesture model.
Three-dimension gesture model parameter vector is down to 12 dimensions from 27 dimensions by S108.
S109 carries out gesture identification according to the three-dimension gesture model after dimensionality reduction.
In conclusion three-dimensional static gesture identification method according to an embodiment of the present invention, by obtaining gesture direct picture
With gesture side image, and generate three-dimension gesture model, dimension-reduction treatment then carried out to three-dimension gesture model, and according to dimensionality reduction after
Three-dimension gesture model carry out gesture identification, the precision and success rate of gesture identification can not only be improved, additionally it is possible to reduce gesture
The operation time of identification and power consumption.
The three-dimensional static gesture identification method of corresponding above-described embodiment, the present invention also propose a kind of three-dimensional static gesture identification
Device.
As shown in fig. 6, the three-dimensional static gesture identifying device of the embodiment of the present invention, including the first image acquisition units 1,
Two image acquisition units 2 and terminal 4.Wherein, the first image acquisition units 1 are for acquiring gesture direct picture；Second
Image acquisition units 2 are for acquiring gesture side image；Terminal 4 respectively with the first image acquisition units 1 and the second figure
Picture acquisition unit 2 is connected by communication interface, and terminal 4 is used to be generated according to gesture direct picture and gesture side image
Three-dimension gesture model, and dimension-reduction treatment is carried out to three-dimension gesture model, and hand is carried out according to the three-dimension gesture model after dimensionality reduction
Gesture identification.
In one embodiment of the invention, communication interface is USB data transfer interface or api interface.
In one embodiment of the invention, terminal 4 includes master control and the human-computer interaction module being connected with master control,
Human-computer interaction module can be used for being arranged the value of above-mentioned N, check gesture identification result etc..
The more specific embodiment of three-dimensional static gesture identifying device can refer to above-mentioned three-dimensional static gesture identification method
Embodiment, details are not described herein.
Three-dimensional static gesture identifying device according to an embodiment of the present invention is acquiring gesture just by the first image acquisition units
Face image acquires gesture side image by the second image acquisition units, by terminal according to gesture direct picture and
Gesture side image generates three-dimension gesture model, and carries out dimension-reduction treatment to three-dimension gesture model, and according to three after dimensionality reduction
It ties up gesture model and carries out gesture identification, the precision and success rate of gesture identification can not only be improved, additionally it is possible to reduce gesture identification
Operation time and power consumption.
In the description of the present invention, the meaning of " plurality " is two or more, unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected；It can be directly connected, can also can be in two elements indirectly connected through an intermediary
The interaction relationship of the connection in portion or two elements.It for the ordinary skill in the art, can be according to specific feelings
Condition understands the concrete meaning of above-mentioned term in the present invention.
In the present invention unless specifically defined or limited otherwise, fisrt feature in the second feature " on " or " down " can be with
It is that the first and second features directly contact or the first and second features pass through intermediary mediate contact.Moreover, fisrt feature exists
Second feature " on ", " top " and " above " but fisrt feature be directly above or diagonally above the second feature, or be merely representative of
First feature horizontal height is higher than second feature.Fisrt feature can be under the second feature " below ", " below " and " below "
One feature is directly under or diagonally below the second feature, or is merely representative of first feature horizontal height less than second feature.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
It although an embodiment of the present invention has been shown and described, for the ordinary skill in the art, can be with
A variety of variations, modification, replacement can be carried out to these embodiments without departing from the principles and spirit of the present invention by understanding
And modification, the scope of the present invention is defined by the appended.
Claims (10)
1. a kind of three-dimensional static gesture identification method characterized by comprising
Obtain gesture direct picture；
Obtain gesture side image；
Three-dimension gesture model is generated according to the gesture direct picture and the gesture side image；
Dimension-reduction treatment is carried out to the three-dimension gesture model；
Gesture identification is carried out according to the three-dimension gesture model after dimensionality reduction.
2. three-dimensional static gesture identification method according to claim 1, which is characterized in that pass through corresponding target hand front
The first image acquisition units being arranged acquire the gesture direct picture, and pass through the guide rail along corresponding target hand side setting
Second image acquisition units of movement are from multiple and different multiple gesture side images of angle acquisition.
3. three-dimensional static gesture identification method according to claim 2, which is characterized in that the guide rail is arc-shaped guide rail,
Plane where the arc-shaped guide rail is perpendicular to the plane where the palm of the target hand.
4. three-dimensional static gesture identification method according to claim 3, which is characterized in that the first image acquisition unit
The primary gesture direct picture is acquired, second image acquisition units move to second from first end along the arc-shaped guide rail
End, which acquires multiple gesture side images or moves to first end from second end, acquires multiple gesture side images, completes gesture front
The synchronous acquisition of image and gesture side image.
5. three-dimensional static gesture identification method according to any one of claim 1-3, which is characterized in that according to the hand
Gesture direct picture and the gesture side image generate three-dimension gesture model, specifically include:
With reference to the constraint relationship of three-dimensional hand skeleton model, three-dimension gesture model finger joint ginseng is obtained according to the gesture direct picture
Number；
The area of hand in each gesture side image is calculated, and the smallest gesture side image of the area for choosing hand is made
To select gesture side image；
In conjunction with three-dimension gesture model finger joint parameter, according to the selected gesture side image obtain three-dimension gesture model end finger joint with
The angle parameter of the plane of delineation；
In conjunction with the angle parameter of the three-dimension gesture model finger joint parameter and three-dimension gesture model end finger joint and the plane of delineation
Generate the three-dimension gesture model.
6. three-dimensional static gesture identification method according to claim 5, which is characterized in that the three-dimension gesture model finger joint
Parameter includes the length and root finger joint, middle finger joint, the angle for holding finger joint and the plane of delineation of each finger finger joint.
7. three-dimensional static gesture identification method according to claim 6, which is characterized in that the three-dimension gesture model into
Row dimension-reduction treatment, specifically includes:
Three-dimension gesture model parameter vector is down to 12 dimensions from 27 dimensions.
8. three-dimensional static gesture identification method according to claim 7, which is characterized in that according to the three-dimension gesture after dimensionality reduction
Model carries out gesture identification, specifically includes:
It is input with the three-dimension gesture model after dimensionality reduction, obtains three-dimensional hand with the gesture identification deep learning model pre-established
Potential model feature vector；
The three-dimension gesture model feature vector is quantified, obtains single gesture discrete features vector, and according to the list
A gesture discrete features vector is trained, and obtains static gesture classifier parameters；
Static gesture classifier after training is used to carry out the single gesture discrete features vector classification judgement, exports hand
Gesture recognition result.
9. a kind of three-dimensional static gesture identifying device characterized by comprising
First image acquisition units, the first image acquisition unit is for acquiring gesture direct picture；
Second image acquisition units, second image acquisition units are for acquiring gesture side image；
Terminal, the terminal respectively with the first image acquisition unit and second image acquisition units
It is connected by communication interface, the terminal is used to be generated according to the gesture direct picture and the gesture side image
Three-dimension gesture model, and to the three-dimension gesture model carry out dimension-reduction treatment, and according to the three-dimension gesture model after dimensionality reduction into
Row gesture identification.
10. three-dimensional static gesture identifying device according to claim 9, which is characterized in that the communication interface is USB number
According to coffret or api interface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811135368.3A CN109409236B (en) | 2018-09-28 | 2018-09-28 | Three-dimensional static gesture recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811135368.3A CN109409236B (en) | 2018-09-28 | 2018-09-28 | Three-dimensional static gesture recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109409236A true CN109409236A (en) | 2019-03-01 |
| CN109409236B CN109409236B (en) | 2020-12-08 |
Family
ID=65465408
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811135368.3A Active CN109409236B (en) | 2018-09-28 | 2018-09-28 | Three-dimensional static gesture recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409236B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111123986A (en) * | 2019-12-25 | 2020-05-08 | 四川云盾光电科技有限公司 | Control device for controlling two-degree-of-freedom turntable based on gestures |
| WO2021000327A1 (en) * | 2019-07-04 | 2021-01-07 | 深圳市瑞立视多媒体科技有限公司 | Hand model generation method, apparatus, terminal device, and hand motion capture method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101383055A (en) * | 2008-09-18 | 2009-03-11 | 北京中星微电子有限公司 | Three-dimensional human face constructing method and system |
| CN102034097A (en) * | 2010-12-21 | 2011-04-27 | 中国科学院半导体研究所 | Method for recognizing human face by comprehensively utilizing front and lateral images |
| US20130088426A1 (en) * | 2010-06-15 | 2013-04-11 | Osamu Shigeta | Gesture recognition device, gesture recognition method, and program |
| CN107688391A (en) * | 2017-09-01 | 2018-02-13 | 广州大学 | A kind of gesture identification method and device based on monocular vision |
-
2018
- 2018-09-28 CN CN201811135368.3A patent/CN109409236B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101383055A (en) * | 2008-09-18 | 2009-03-11 | 北京中星微电子有限公司 | Three-dimensional human face constructing method and system |
| US20130088426A1 (en) * | 2010-06-15 | 2013-04-11 | Osamu Shigeta | Gesture recognition device, gesture recognition method, and program |
| CN102034097A (en) * | 2010-12-21 | 2011-04-27 | 中国科学院半导体研究所 | Method for recognizing human face by comprehensively utilizing front and lateral images |
| CN107688391A (en) * | 2017-09-01 | 2018-02-13 | 广州大学 | A kind of gesture identification method and device based on monocular vision |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021000327A1 (en) * | 2019-07-04 | 2021-01-07 | 深圳市瑞立视多媒体科技有限公司 | Hand model generation method, apparatus, terminal device, and hand motion capture method |
| CN111123986A (en) * | 2019-12-25 | 2020-05-08 | 四川云盾光电科技有限公司 | Control device for controlling two-degree-of-freedom turntable based on gestures |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109409236B (en) | 2020-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kuch et al. | Vision based hand modeling and tracking for virtual teleconferencing and telecollaboration | |
| Erol et al. | Vision-based hand pose estimation: A review | |
| Xu et al. | A review: Point cloud-based 3d human joints estimation | |
| CN113362452B (en) | Hand posture three-dimensional reconstruction method and device and storage medium | |
| CN104035557B (en) | Kinect action identification method based on joint activeness | |
| CN101833788A (en) | A 3D human body modeling method using hand-drawn sketches | |
| CN105045496B (en) | A kind of gesture interaction method based on joint point transformation | |
| WO2021000327A1 (en) | Hand model generation method, apparatus, terminal device, and hand motion capture method | |
| CN110490959A (en) | Three dimensional image processing method and device, virtual image generation method and electronic equipment | |
| CN109960403A (en) | Visual presentation and interaction methods for medical images in an immersive environment | |
| CN109409236A (en) | Three-dimensional static gesture identification method and device | |
| CN107633551A (en) | A display method and device for a virtual keyboard | |
| CN110866468A (en) | A gesture recognition system and method based on passive RFID | |
| He et al. | A New Kinect‐Based Posture Recognition Method in Physical Sports Training Based on Urban Data | |
| CN112183316B (en) | Athlete human body posture measuring method | |
| CN208569551U (en) | A learnable data acquisition system based on gesture recognition gloves | |
| Jabalameli et al. | From single 2D depth image to gripper 6D pose estimation: A fast and robust algorithm for grabbing objects in cluttered scenes | |
| Prasad et al. | A wireless dynamic gesture user interface for HCI using hand data glove | |
| Wang et al. | Design of a four-axis robot arm system based on machine vision | |
| CN113569775B (en) | Mobile terminal real-time 3D human motion capturing method and system based on monocular RGB input, electronic equipment and storage medium | |
| CN109032355A (en) | Various gestures correspond to the flexible mapping interactive algorithm of same interactive command | |
| Yan et al. | AGRMTS: A virtual aircraft maintenance training system using gesture recognition based on PSO‐BPNN model | |
| CN110517338A (en) | A method of reusable maneuver library is constructed based on two sufficient role's substance features | |
| Chien et al. | Robotic calligraphy system using delta-like robot manipulator and virtual brush model | |
| Ugolotti et al. | Differential evolution based human body pose estimation from point clouds |
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 |