CN112987926A - Method and system for reconstructing volley gesture track - Google Patents
Method and system for reconstructing volley gesture track Download PDFInfo
- Publication number
- CN112987926A CN112987926A CN202110245043.6A CN202110245043A CN112987926A CN 112987926 A CN112987926 A CN 112987926A CN 202110245043 A CN202110245043 A CN 202110245043A CN 112987926 A CN112987926 A CN 112987926A
- Authority
- CN
- China
- Prior art keywords
- gesture
- volley
- data
- motion data
- axis motion
- 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
Images
Classifications
-
- 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
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
- G06F3/014—Hand-worn input/output arrangements, e.g. data gloves
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
- G06F2218/06—Denoising by applying a scale-space analysis, e.g. using wavelet analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a volley gesture track reconstruction method and a volley gesture track reconstruction system, wherein the method comprises the following steps: acquiring six-axis motion data of a gesture captured by a six-axis sensor, and determining whether the six-axis motion data belongs to a preset gesture motion library, wherein the six-axis motion data comprises acceleration data and gyroscope data; when the six-axis motion data are determined to belong to a preset gesture action library, carrying out noise reduction and filtering pretreatment on the six-axis motion data by adopting a Mallat algorithm to obtain pretreated six-axis motion data; performing feature extraction on the preprocessed six-axis motion data to obtain gesture motion features; according to the gesture motion characteristics, the track size of the gesture motion is reconstructed, and the method has higher stability and reliability in coarse-grained volley gesture track size reconstruction.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a volley gesture track reconstruction method and system.
Background
The ubiquitous environment is an environment with computing and communication capabilities, wherein an information space and a physical space are fused. In a general environment, the calculation is centered on people, man-machine interaction is similar to a natural communication mode between people, equipment for calculation is ubiquitous, the equipment is integrated into the living environment of people, and required services can be conveniently provided for people.
The volley gestures are non-contact air gestures which enable a user to operate in a free-hand mode, and are essentially a natural man-machine interaction mode which does not bring any inconvenience to the user gesture interaction. For the reconstruction of the volley gesture track, in the prior art, a strapdown inertial navigation algorithm is generally adopted to reconstruct the motion track determined by the MPU9250 sensor. This solution has the following drawbacks:
the strapdown inertial navigation algorithm mainly utilizes data collected by a three-axis accelerometer, a three-axis gyroscope and a three-axis magnetometer in an MPU6050 or an MPU9250 to perform operation, firstly, acceleration data under a carrier coordinate system (a carrier is an inertial sensor on a hand) is converted into acceleration data under a world coordinate system, referring to fig. 1, in the world coordinate system, an X axis points to a geographical east, a Y axis points to geographical north, a Z axis points to the sky, and the three axes are vertical to each other; and then calculating acceleration data of an X axis, a Y axis and a Z axis in a world coordinate system, and performing secondary integration on the acceleration data to obtain the displacement of the volley gesture in the air. However, due to the influence of various factors, the measured displacement is noisy, and particularly for inertial sensor equipment with low precision, large displacement deviation still exists even if various processing methods are adopted, so that reconstruction of the size of the volley gesture track under the head-up view angle based on the method is extremely unstable and unreliable.
Therefore, a scheme is urgently needed to be provided to solve the problem of instability of the strapdown inertial navigation algorithm to the low-precision inertial sensor-based volley gesture track reconstruction.
Disclosure of Invention
The present invention is directed to a method and a system for reconstructing a volley gesture trajectory, so as to solve one or more technical problems in the prior art and provide at least one useful choice or creation condition.
In order to achieve the purpose, the invention provides the following technical scheme:
a volley gesture trajectory reconstruction method, the method comprising the steps of:
acquiring six-axis motion data of a gesture captured by a six-axis sensor, and determining whether the six-axis motion data belongs to a preset gesture motion library, wherein the six-axis motion data comprises acceleration data and gyroscope data;
when the six-axis motion data are determined to belong to a preset gesture action library, carrying out noise reduction and filtering pretreatment on the six-axis motion data by adopting a Mallat algorithm to obtain pretreated six-axis motion data;
performing feature extraction on the preprocessed six-axis motion data to obtain gesture motion features;
and reconstructing the track size of the gesture action according to the gesture motion characteristics.
Further, the six-axis sensor is a low-precision sensor MPU 6050.
Further, the performing noise reduction and filtering preprocessing on the six-axis motion data by using a Mallat algorithm to obtain preprocessed six-axis motion data includes:
determining wavelet basis and wavelet decomposition level number N, and performing wavelet decomposition on the six-axis motion data to obtain wavelet coefficients of each level;
determining the threshold of each level of wavelet coefficient through a threshold selection criterion, and respectively acting the threshold of each level of wavelet coefficient on each level of wavelet coefficient based on a threshold function to obtain the wavelet coefficient subjected to threshold quantization;
and (4) performing inverse transformation on the wavelet coefficient subjected to the threshold quantization action to obtain the preprocessed six-axis motion data.
Further, the threshold of the wavelet coefficients of each level is determined by multiple tests, the threshold function is a hard threshold function, and the calculation formula of the threshold function is as follows:
Further, the feature extraction is performed on the six-axis motion data after the preprocessing, so as to obtain gesture motion features, and the method comprises the following steps:
carrying out normalization processing on the preprocessed six-axis motion data to obtain normalized data;
and extracting gesture motion characteristics from the normalized data, wherein the gesture motion characteristics comprise time domain characteristics and frequency domain characteristics.
Further, the time domain feature includes at least one of the following feature information: maximum, minimum, mean, variance in the time domain;
the frequency domain features include at least one of the following feature information: mean frequency, root mean square frequency, peak frequency, sum frequency, standard deviation, variance in the frequency domain.
Further, reconstructing the trajectory size of the gesture motion according to the gesture motion characteristics specifically includes:
inputting the training set into a machine learning model for training to obtain a gesture recognition model; the training set comprises pre-acquired gesture motion characteristic samples;
and inputting the gesture motion characteristics into the gesture recognition model, and finishing track size reconstruction according to the obtained size of the volley track.
A computer readable storage medium having stored thereon a volley gesture trajectory reconstruction program which, when executed by a processor, implements the steps of the volley gesture trajectory reconstruction method as recited in any one of the above.
A volley gesture trajectory reconstruction system, the system comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement any of the above volley gesture trajectory reconstruction methods.
The invention has the beneficial effects that: the invention discloses a volley gesture track reconstruction method and a volley gesture track reconstruction system, six-axis motion data are captured through a six-axis sensor of a low-precision inertial sensor, and the six-axis motion data are subjected to noise reduction and filtering preprocessing by adopting a Mallat algorithm to obtain preprocessed six-axis motion data; the wavelet threshold denoising method based on the Mallat algorithm has the advantages of low entropy, multi-resolution, decorrelation, flexibility and the like, finally feature extraction is carried out on the preprocessed six-axis motion data to obtain gesture motion features, and the track size of gesture motion is reconstructed according to the gesture motion features. The method has higher stability and reliability in coarse-grained volley gesture track size reconstruction.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a volley gesture trajectory reconstruction method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of wavelet decomposition of the six-axis motion data based on the Mallat algorithm in the embodiment of the present invention.
Detailed Description
The conception, specific structure and technical effects of the present application will be described clearly and completely with reference to the following embodiments and the accompanying drawings, so that the purpose, scheme and effects of the present application can be fully understood. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Referring to fig. 1, fig. 1 shows a volley gesture trajectory reconstruction method provided by an embodiment of the present application, where the method includes the following steps:
s100, acquiring six-axis motion data of a gesture captured by a six-axis sensor, and determining whether the six-axis motion data belong to a preset gesture motion library, wherein the six-axis motion data comprise acceleration data and gyroscope data;
s200, when the six-axis motion data belong to a preset gesture action library, performing noise reduction and filtering pretreatment on the six-axis motion data by adopting a Mallat algorithm to obtain pretreated six-axis motion data;
step S300, performing feature extraction on the preprocessed six-axis motion data to obtain gesture motion features;
and S400, reconstructing the track size of the gesture motion according to the gesture motion characteristics.
The gesture motion library is used for reconstructing the trajectory size of the gesture in the air, and the preset gesture motion library refers to the same preset gesture but different in gesture amplitude. In some exemplary embodiments, the gesture motion is set as a horizontal line motion, the horizontal line motion with the amplitude of 10cm, 20cm, 30cm, 40cm, 50cm and 60cm is performed respectively (the gesture shown here is only an example, and the method is not only applicable to such gestures), then in the motion process, the six-axis sensor generates motion data according to the captured gesture motion, and sends the motion data to an upper computer (a terminal with a data processing function) at a set frequency for processing, and finally, the trajectory size of the gesture motion is reconstructed. The gesture motion in step S400 is a gesture motion determined to be preset according to the six-axis motion data.
In a preferred embodiment, the six-axis sensor is a low-precision sensor MPU 6050.
As shown in fig. 2, in a preferred embodiment, in step S200, the performing denoising and filtering preprocessing on the six-axis motion data by using the Mallat algorithm to obtain preprocessed six-axis motion data includes:
step S210, determining wavelet basis and wavelet decomposition level number N, and performing wavelet decomposition on the six-axis motion data to obtain wavelet coefficients of each level;
in this step, the Wavelet basis is preferably a multi-bezier Wavelet basis (Daubechies-Wavelet), which has good regularity and orthogonality, and the number of vanishing moments increases with the increase of the number of decomposition levels.
The wavelet basis and the wavelet decomposition level number N are determined by a combination of qualitative analysis and quantitative analysis, in some embodiments, a wavelet filter is used to perform wavelet decomposition on the six-axis motion data, the number of layers is selected to be four-layer decomposition, fig. 2 is a schematic structural form of performing three-level wavelet decomposition on the six-axis motion data based on Mal lat algorithm, the six-axis motion data is decomposed into low-frequency coefficients and high-frequency coefficients based on Mal lat algorithm, and there are S1 + d2+ d3+ d4+ a4, where S represents the six-axis motion data, a1, a2, a3, a4 all represent low-frequency coefficients, and d1, d2, d3, d4 all represent high-frequency coefficients.
Step S220, determining the threshold of each level of wavelet coefficient through a threshold selection criterion, and respectively applying the threshold of each level of wavelet coefficient to each level of wavelet coefficient based on a threshold function to obtain the wavelet coefficient subjected to threshold quantization;
and step S230, performing inverse transformation on the wavelet coefficient subjected to the threshold quantization function to obtain preprocessed six-axis motion data.
In the embodiment provided by the invention, the threshold value and the threshold value function of each level of wavelet coefficient are required to be selected, the wavelet basis is determined through the threshold value, and the decomposition level number is determined through the threshold value function. In a preferred embodiment, the threshold of each wavelet coefficient is determined by multiple tests, the threshold function is a hard threshold function, and the calculation formula of the threshold function is as follows:
The invention applies the Mallat algorithm to carry out denoising and filtering preprocessing, and the Mallat algorithm unifies the wavelet transformation theory so that the wavelet transformation theory becomes feasible in calculation. The wavelet threshold denoising method can better separate noise and useful signals in wavelet coefficients, and comprises the steps of respectively processing parts of wavelet coefficients larger than or smaller than a certain threshold in each level of coefficients obtained after wavelet decomposition, and then performing wavelet reconstruction to obtain denoised signals. Because the output data of the MEMS sensor is non-stationary and non-linear, the invention adopts the noise reduction method to carry out noise reduction processing on the six-axis motion data, compared with the traditional filtering and noise reduction method, the wavelet threshold noise reduction based on the Mallat algorithm has the advantages of low entropy, multi-resolution, decorrelation, flexibility and the like, and has great advantages in the aspects of signal noise reduction and processing of the MEMS sensor.
In a preferred embodiment, the step S300 includes:
step S310, carrying out normalization processing on the preprocessed six-axis motion data to obtain normalized data;
preferably, the normalization is performed by using a dispersion normalization (min-max-normalization), and the data after the normalization process is between (0, 1), so that the subsequent feature extraction is easy to perform. The normalized formula is:
wherein X is any kind of data in the six-axis motion data, Xmin is the minimum value of the kind of data, Xmax is the maximum value of the kind of data,
in order to normalize the data after the data normalization process, in this embodiment, each piece of data is processed in this manner.
Step S320, extracting gesture motion characteristics from the normalized data, wherein the gesture motion characteristics comprise time domain characteristics and frequency domain characteristics.
The extraction of the gesture motion features refers to selecting feature values according to the characteristics of data for classification selection of a subsequent machine learning model. In the embodiment of the present invention, using ten features of the normalized data, as shown in fig. 2, the normalized data is extracted in the time domain and the frequency domain, respectively, where the time domain refers to a time series signal, and the frequency domain refers to a frequency signal obtained by fourier transforming the time domain signal.
In a preferred embodiment, the time domain feature comprises at least one of the following feature information: maximum, minimum, mean, variance in the time domain;
the frequency domain features include at least one of the following feature information: mean frequency, root mean square frequency, peak frequency, sum frequency, standard deviation, variance in the frequency domain.
In a preferred embodiment, the step S400 specifically includes:
inputting the training set into a machine learning model for training to obtain a gesture recognition model; the training set comprises pre-acquired gesture motion characteristic samples;
and inputting the gesture motion characteristics into the gesture recognition model, and finishing track size reconstruction according to the obtained size of the volley track.
In order to solve the problem of instability of reconstruction of the flight gesture track based on a low-precision inertial sensor in the prior art, the coarse-grained track reconstruction method utilizes a machine learning model (such as a support vector machine, a decision tree, a random forest and other models) to carry out coarse-grained track reconstruction on the flight gesture track, so that the track reconstruction has higher stability and reliability; in the embodiment, a plurality of different gesture motion characteristic samples are imported into a machine learning model for training, so that the model can learn different sizes of the trajectories of the volley gestures; the number of the gesture motion characteristic samples is set according to actual conditions, and it can be understood that when the number of the gesture motion characteristic samples is larger, the obtained gesture recognition model is more accurate.
In some exemplary embodiments, the gesture motion is set as horizontal traverse motions, and horizontal traverse motions with amplitudes of 10cm, 20cm, 30cm, 40cm, 50cm and 60cm are respectively made (the gesture shown here is only an example, and the method is not only applicable to such gestures), and the horizontal traverse dimensions of 10cm, 20cm, 30cm, 40cm, 50cm and 60cm can be effectively resolved through a trained model through experimental verification. According to the method, the size of the trajectory of the gesture in the air can be stably identified, illustratively, the gesture displaces 18cm in the air, the trajectory size of the volley gesture is identified to be any value from 16cm to 20cm through the method disclosed by the embodiment, the method disclosed by the invention belongs to coarse-grained volley gesture trajectory size reconstruction, and the method has higher stability, feasibility and reliability.
Corresponding to the method in fig. 1, an embodiment of the present invention further provides a computer-readable storage medium, where a volley gesture trajectory reconstruction program is stored on the computer-readable storage medium, and when executed by a processor, the volley gesture trajectory reconstruction program implements the steps of the volley gesture trajectory reconstruction method according to any one of the above embodiments.
Corresponding to the method in fig. 1, an embodiment of the present invention further provides a volley gesture trajectory reconstruction system, where the system includes:
at least one processor;
at least one memory for storing at least one program;
when the at least one program is executed by the at least one processor, the at least one processor is enabled to implement the volley gesture trajectory reconstruction method according to any one of the above embodiments.
The contents in the above method embodiments are all applicable to the present system embodiment, the functions specifically implemented by the present system embodiment are the same as those in the above method embodiment, and the beneficial effects achieved by the present system embodiment are also the same as those achieved by the above method embodiment.
The Processor may be a Central-Processing Unit (CPU), other general-purpose Processor, a Digital Signal Processor (DSP), an Application-Specific-Integrated-Circuit (ASIC), a Field-Programmable Gate array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor, etc., and the processor is a control center of the volley gesture trajectory reconstruction system and connects various parts of the whole apparatus operable by the volley gesture trajectory reconstruction system by using various interfaces and lines.
The memory may be used to store the computer programs and/or modules, and the processor may implement the various functions of the volley gesture trajectory reconstruction system by running or executing the computer programs and/or modules stored in the memory and invoking the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart-Media-Card (SMC), a Secure-Digital (SD) Card, a Flash-memory Card (Flash-Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
While the description of the present application has been made in considerable detail and with particular reference to a few illustrated embodiments, it is not intended to be limited to any such details or embodiments or any particular embodiments, but it is to be construed that the present application effectively covers the intended scope of the application by reference to the appended claims, which are interpreted in view of the broad potential of the prior art. Further, the foregoing describes the present application in terms of embodiments foreseen by the inventor for which an enabling description was available, notwithstanding that insubstantial changes from the present application, not presently foreseen, may nonetheless represent equivalents thereto.
Claims (9)
1. A volley gesture trajectory reconstruction method is characterized by comprising the following steps:
acquiring six-axis motion data of a gesture captured by a six-axis sensor, and determining whether the six-axis motion data belongs to a preset gesture motion library, wherein the six-axis motion data comprises acceleration data and gyroscope data;
when the six-axis motion data are determined to belong to a preset gesture action library, carrying out noise reduction and filtering pretreatment on the six-axis motion data by adopting a Mallat algorithm to obtain pretreated six-axis motion data;
performing feature extraction on the preprocessed six-axis motion data to obtain gesture motion features;
and reconstructing the track size of the gesture action according to the gesture motion characteristics.
2. The volley gesture trajectory reconstruction method according to claim 1, wherein the six-axis sensor is a low-precision sensor MPU 6050.
3. The volley gesture trajectory reconstruction method according to claim 2, wherein the preprocessing of denoising and filtering the six-axis motion data by using a Mallat algorithm to obtain the preprocessed six-axis motion data comprises:
determining wavelet basis and wavelet decomposition level number N, and performing wavelet decomposition on the six-axis motion data to obtain wavelet coefficients of each level;
determining the threshold of each level of wavelet coefficient through a threshold selection criterion, and respectively acting the threshold of each level of wavelet coefficient on each level of wavelet coefficient based on a threshold function to obtain the wavelet coefficient subjected to threshold quantization;
and (4) performing inverse transformation on the wavelet coefficient subjected to the threshold quantization action to obtain the preprocessed six-axis motion data.
4. The volley gesture trajectory reconstruction method according to claim 3, wherein the threshold of each level of wavelet coefficients is determined by multiple tests, the threshold function is a hard threshold function, and the calculation formula of the threshold function is as follows:
5. The volley gesture trajectory reconstruction method according to claim 4, wherein the performing feature extraction on the preprocessed six-axis motion data to obtain the gesture motion features comprises:
carrying out normalization processing on the preprocessed six-axis motion data to obtain normalized data;
and extracting gesture motion characteristics from the normalized data, wherein the gesture motion characteristics comprise time domain characteristics and frequency domain characteristics.
6. The volley gesture trajectory reconstruction method according to claim 5, wherein the time domain feature includes at least one of the following feature information: maximum, minimum, mean, variance in the time domain;
the frequency domain features include at least one of the following feature information: mean frequency, root mean square frequency, peak frequency, sum frequency, standard deviation, variance in the frequency domain.
7. The volley gesture trajectory reconstruction method according to claim 6, wherein reconstructing the trajectory size of the gesture motion according to the gesture motion characteristics specifically includes:
inputting the training set into a machine learning model for training to obtain a gesture recognition model; the training set comprises pre-acquired gesture motion characteristic samples;
and inputting the gesture motion characteristics into the gesture recognition model, and finishing track size reconstruction according to the obtained size of the volley track.
8. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, carries out the steps of the volley gesture trajectory reconstruction method according to one of claims 1 to 7.
9. A volley gesture trajectory reconstruction system, the system comprising:
at least one processor;
at least one memory for storing at least one program;
when executed by the at least one processor, cause the at least one processor to implement the volley gesture trajectory reconstruction method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245043.6A CN112987926B (en) | 2021-03-05 | 2021-03-05 | Method and system for reconstructing volley gesture track |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245043.6A CN112987926B (en) | 2021-03-05 | 2021-03-05 | Method and system for reconstructing volley gesture track |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112987926A true CN112987926A (en) | 2021-06-18 |
| CN112987926B CN112987926B (en) | 2023-01-24 |
Family
ID=76353040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110245043.6A Active CN112987926B (en) | 2021-03-05 | 2021-03-05 | Method and system for reconstructing volley gesture track |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112987926B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100033352A1 (en) * | 2008-08-08 | 2010-02-11 | Industrial Technology Research Institute | Real-time motion recognition method and inertia-sensing and trajectory-reconstruction device using the same |
| CN101655743A (en) * | 2008-08-18 | 2010-02-24 | 财团法人工业技术研究院 | Instant action identifying method and inertia sensing and track reconstructing device thereof |
| US20160092504A1 (en) * | 2014-09-27 | 2016-03-31 | Intel Corporation | Recognition of Free-form Gestures from Orientation Tracking of a Handheld or Wearable Device |
| CN110163142A (en) * | 2019-05-17 | 2019-08-23 | 重庆大学 | Real-time gesture recognition method and system |
| US20200217930A1 (en) * | 2018-08-30 | 2020-07-09 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Method for gesture recognition, terminal, and storage medium |
-
2021
- 2021-03-05 CN CN202110245043.6A patent/CN112987926B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100033352A1 (en) * | 2008-08-08 | 2010-02-11 | Industrial Technology Research Institute | Real-time motion recognition method and inertia-sensing and trajectory-reconstruction device using the same |
| CN101655743A (en) * | 2008-08-18 | 2010-02-24 | 财团法人工业技术研究院 | Instant action identifying method and inertia sensing and track reconstructing device thereof |
| US20160092504A1 (en) * | 2014-09-27 | 2016-03-31 | Intel Corporation | Recognition of Free-form Gestures from Orientation Tracking of a Handheld or Wearable Device |
| US20200217930A1 (en) * | 2018-08-30 | 2020-07-09 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Method for gesture recognition, terminal, and storage medium |
| CN110163142A (en) * | 2019-05-17 | 2019-08-23 | 重庆大学 | Real-time gesture recognition method and system |
Non-Patent Citations (1)
| Title |
|---|
| 刘瑞英: "基于小波分析的轨迹跟踪系统的研究", 《中国优秀博硕士学位论文全文数据库（硕士）信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112987926B (en) | 2023-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108351984B (en) | Hardware-efficient deep convolutional neural network | |
| CN110797021B (en) | Hybrid speech recognition network training method, hybrid speech recognition device and storage medium | |
| CN108205803B (en) | Image processing method, and training method and device of neural network model | |
| Ha et al. | Multi-modal convolutional neural networks for activity recognition | |
| CN103984416B (en) | Gesture recognition method based on acceleration sensor | |
| CN110222700A (en) | SAR image recognition methods and device based on Analysis On Multi-scale Features and width study | |
| CN105159463B (en) | A kind of contactless wearable intelligent ring system and its gesture identification method | |
| CN111144566B (en) | Training method for neural network weight parameters, feature classification method and corresponding device | |
| CN107392123B (en) | Radio frequency fingerprint feature extraction and identification method based on coherent accumulation noise elimination | |
| CN109426828B (en) | Point cloud classification method, device, equipment and storage medium | |
| CN112633420B (en) | Image similarity determination and model training method, device, equipment and medium | |
| WO2022094293A1 (en) | Deep-learning based speech enhancement | |
| He | Accelerometer Based Gesture Recognition Using Fusion Features and SVM. | |
| CN109766798A (en) | Gesture data processing method, server and awareness apparatus based on experience small echo | |
| CN113837942A (en) | Super-resolution image generation method, device, equipment and storage medium based on SRGAN | |
| CN112364738A (en) | Human body posture estimation method, device, system and medium based on deep learning | |
| CN112987926B (en) | Method and system for reconstructing volley gesture track | |
| CN115171713A (en) | Voice noise reduction method, device and equipment and computer readable storage medium | |
| CN102855873A (en) | Electronic equipment and method used for controlling same | |
| CN117099136A (en) | Dynamic head for object detection | |
| CN107341519B (en) | Support vector machine identification optimization method based on multi-resolution analysis | |
| WO2022205249A1 (en) | Audio feature compensation method, audio recognition method, and related product | |
| CN115329820A (en) | Power grid disturbance identification method and system based on improved FastICA | |
| CN107992193A (en) | Gesture confirmation method, device and electronic equipment | |
| Bashir et al. | The impact of feature vector length on activity recognition accuracy on mobile phone |
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 |