CN114167978A - Human-computer interaction system carried on construction robot - Google Patents
Human-computer interaction system carried on construction robot Download PDFInfo
- Publication number
- CN114167978A CN114167978A CN202111335163.1A CN202111335163A CN114167978A CN 114167978 A CN114167978 A CN 114167978A CN 202111335163 A CN202111335163 A CN 202111335163A CN 114167978 A CN114167978 A CN 114167978A
- Authority
- CN
- China
- Prior art keywords
- construction
- gesture
- frame
- worker
- human
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 101
- 230000003993 interaction Effects 0.000 title claims abstract description 28
- 230000009471 action Effects 0.000 claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 10
- 230000011664 signaling Effects 0.000 claims 1
- 238000009435 building construction Methods 0.000 abstract description 2
- 238000013135 deep learning Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
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
- G06F3/017—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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
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)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- User Interface Of Digital Computer (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a man-machine interaction system carried on a construction robot, which comprises the following parts: the position tracking module is used for detecting and tracking the position of a construction worker; the gesture tracking module is used for tracking and identifying hand actions of construction workers when the construction workers are detected to make gestures to the robot; and the gesture recognition module is used for recognizing the meaning of the gesture of the worker and outputting a corresponding instruction. The method completes gesture recognition by collecting construction workers and applying a deep learning method, can receive gesture instructions of the workers on the premise of not influencing the work of the construction workers, has good universality and effectiveness, and is widely applied to the field of building construction.
Description
Technical Field
The invention relates to the field of computer vision, in particular to a man-machine interaction system carried on a construction robot.
Background
In building construction, life and property loss is caused by irregular operation and various accidents, and construction efficiency is reduced. In order to improve the safety of the efficiency of building production, the construction robot is more and more widely applied to the construction site. However, the construction robot cannot communicate with human directly, and therefore, communication between construction workers and the construction robot needs to be achieved through various human-computer interaction technologies.
Common human-computer interaction technologies include operation of a rocker and a controller, but the technologies all require manual operation of a construction worker to realize interaction, and human-computer interaction cannot be realized while the construction worker works. The construction robot is also provided with a sensor to track the construction worker, so as to achieve the purpose of man-machine interaction.
Compared with the interaction technology mentioned above, the human-computer interaction technology based on machine vision has obvious advantages. The human-computer interaction is realized through specific actions without needing a construction worker to additionally wear equipment or input instructions. The gesture action has the characteristics of easiness in use, nature and intuition, and is convenient for construction workers and construction robots to learn and use.
Disclosure of Invention
In view of this, embodiments of the present invention provide a human-computer interaction system mounted on a construction robot.
The invention provides a human-computer interaction system mounted on a construction robot, which is characterized by comprising the following parts:
the position tracking module is used for detecting and tracking the position of a construction worker;
the gesture tracking module is used for tracking and identifying hand actions of construction workers when the construction workers are detected to make gestures to the robot;
and the gesture recognition module is used for recognizing the meaning of the gesture of the worker and outputting a corresponding instruction.
Further, the operating steps of the location tracking module include:
the method comprises the steps that image information of construction workers is collected in real time through a construction robot, and a first video sequence is established through collected images;
identifying construction workers in the video sequence, and establishing different identification IDs for each construction worker;
drawing a boundary frame of a construction worker, and modeling appearance information of the construction worker;
and drawing the action track of the construction worker through the video sequence and the appearance information of the construction worker, and associating the collected image with the action track to realize the position tracking of the construction worker.
Further, the location tracking module implements location tracking using a convolutional neural network, which includes YOLOv3 convolutional neural network.
Further, the working steps of the gesture tracking module include:
when receiving a gesture signal of a construction worker, amplifying a first video sequence by taking the construction worker sending the signal as a center, and adjusting the first video sequence according to a boundary frame of the construction worker to obtain a second video sequence;
and performing motion capture on the gestures of the construction workers, and generating and outputting a third video sequence.
Further, the adjusting the first video sequence according to the bounding box of the construction worker specifically includes: and enabling the distance between the boundary box of the construction worker and the edge of the acquired image to be not less than one eighth of the corresponding radial side length of the first video sequence.
Further, the working steps of the gesture recognition module comprise:
detecting a gesture using a detector;
identifying a specific meaning of a gesture with a classifier when the gesture is detected by the detector;
and outputting an operation instruction corresponding to the specific meaning according to the specific meaning of the gesture.
Further, the gesture recognition module is realized through a convolutional neural network based on a hierarchical structure.
Further, the detecting a gesture using a detector specifically includes:
cutting the second video sequence into 8 video frames according to unit time;
carrying out frame-by-frame detection on the video frame, and extracting gesture features in the video frame through a ResNet-10 convolutional neural network;
the video frame in which the gesture was detected is marked as the first frame.
Further, the identifying, by using the classifier, the specific meaning of the gesture specifically includes:
further cropping the video frames, the second video sequence per unit time being cropped into 32 video frames;
establishing a video frame index T, identifying subsequent video frames from a first frame TO, and classifying gestures in the video frames when the difference between T and T0 is equal TO a multiple of a time factor L;
calculating the weighted probability of each frame in the video frame index T through a weighting function, calculating the difference value between the highest value and the next highest value of the weighted probabilities, and searching a corresponding gesture in a library according to the gesture in the video frame when the difference value is larger than a preset threshold value;
and outputting the specific meaning of the corresponding gesture.
Further, the weighting function is formulated as:
wherein, wTReferring to the weight at the Tth frame, u corresponds to the average duration of the gesture (i.e., the number of frames) in the dataset and s is the stride length.
The invention has the beneficial effects that: in an actual construction scene, by means of the man-machine interaction system carried on the construction robot, a construction worker can move and make gestures to the construction robot at the same time, a sensor does not need to be worn, and the working efficiency of the construction worker is guaranteed. The experimental result shows that the method has good overall accuracy and recall rate, and the effectiveness of the method is verified.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a general flow diagram of a human-computer interaction system mounted on a construction robot;
fig. 2 is a flowchart of the operation of a gesture recognition module in a human-computer interaction system mounted on a construction robot.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The embodiment provides a man-machine interaction system carried on a construction robot. And the vision-based convolutional neural network is adopted to capture and explain the gestures of the construction workers so as to guide the operation of the tower crane or other construction equipment.
The system specifically comprises the following modules, and the work flows corresponding to the modules are shown in fig. 1:
the position tracking module is used for detecting and tracking the position of a construction worker;
the gesture tracking module is used for tracking and identifying hand actions of construction workers when the construction workers are detected to make gestures to the robot;
and the gesture recognition module is used for recognizing the meaning of the gesture of the worker and outputting a corresponding instruction.
The present embodiment introduces a location tracking module. The working steps of the position tracking module comprise:
the method comprises the steps that image information of construction workers is collected in real time through a construction robot, and a first video sequence is established through collected images;
identifying construction workers in the video sequence, establishing a different identification I D for each construction worker;
drawing a boundary frame of a construction worker, and modeling appearance information of the construction worker;
and drawing the action track of the construction worker through the video sequence and the appearance information of the construction worker, and associating the collected image with the action track to realize the position tracking of the construction worker.
The purpose of the location tracking module is to extract the construction worker who made the gesture in the video sequence. The detection module identifies the constructors in each frame and obtains their bounding boxes. Given the detection results, the trajectory and appearance information is modeled to correlate the current detection with the existing trajectory to track the worker. When multiple workers are present in the scan, the construction worker making the gesture can be identified by tracking the identification number (I D). In the embodiment, a simple online real-time (sequencing) tracker with multiple object depths based on a YOLOv3 convolutional neural network is used for tracking constructors, the same constructor detected in the previous process is associated to all frames, and the track and appearance information provided by the detection result tracks the constructors in video frames.
The present embodiment introduces a gesture tracking module. The workflow of the gesture tracking module comprises:
when receiving a gesture signal of a construction worker, amplifying a first video sequence by taking the construction worker sending the signal as a center, and adjusting the first video sequence according to a boundary frame of the construction worker to obtain a second video sequence;
and performing motion capture on the gestures of the construction workers, and generating and outputting a third video sequence.
The purpose of the gesture tracking module is to crop the area of the construction worker issuing the gesture from the original frame to form a queue for detecting and classifying gestures. The assembly can be divided into two steps: horizontal extension of the extraction area and formation of a gesture recognition queue. The extracted area is first expanded horizontally by 25% to adequately capture the gestures made by the worker on a trial and error basis. As the worker swings the arm, the area directly obtained by the detection and tracking assembly may miss a portion of the hand area. After extending horizontally to the left and right, the area of the construction worker can capture the entire hand area, which is critical for gesture recognition. The generated gesture recognition queue is used for the work of a subsequent gesture recognition module.
The present embodiment introduces a gesture recognition module. The work flow of the gesture recognition module is shown in fig. 2, and comprises the following steps:
detecting a gesture using a detector;
identifying a specific meaning of the gesture with a classifier when the gesture is detected by the detector;
and outputting an operation instruction corresponding to the specific meaning according to the specific meaning of the gesture.
Wherein, using the detector to detect the gesture specifically includes:
cutting the second video sequence into 8 video frames according to unit time;
carrying out frame-by-frame detection on the video frame, and extracting gesture features in the video frame through a ResNet-10 convolutional neural network;
the video frame in which the gesture is detected is denoted as T0.
This element acts as a switch to determine whether the classifier needs to be activated. If a gesture is detected and the classifier is not activated, the classifier is activated and records the current frame index as the first frame, and TO is the first frame index when the gesture is detected.
The method for recognizing the specific meaning of the gesture by using the classifier specifically comprises the following steps:
further cropping the video frames, the second video sequence per unit time being cropped into 32 video frames;
establishing a video frame index T, identifying subsequent video frames from a first frame TO, and classifying gestures in the video frames when the difference between T and T0 is equal TO a multiple of a time factor L;
calculating the weighted probability of each frame in the video frame index T through a weighting function, calculating the difference value between the highest value and the next highest value of the weighted probabilities, and searching a corresponding gesture in a library according to the gesture in the video frame when the difference value is larger than a preset threshold value;
and outputting the specific meaning of the corresponding gesture.
The formula of the weight function is:
where wT refers to the weight at the tth frame, u corresponds to the average duration of the gesture (i.e., the number of frames) in the dataset, and s is the stride length, which can take the value of 1, which is small enough not to miss any gesture.
With respect to calculating the difference between the highest and second highest weighted probabilities: if the difference is greater than a threshold τ, gesture recognition will be triggered; otherwise, this means that the classifier has insufficient confidence in classifying the gesture type. The architecture will make another round of gesture detection and classification until the detector no longer detects the gesture and disables the classifier. The choice of τ and L depends on the likelihood and frequency with which the user triggers the identification. After repeated experiments, tau and L respectively take values of 0.20 and 15.
Through field experiments, the effectiveness of the implementation theory in gesture recognition is verified, and the overall accuracy and recall rate respectively reach 87.0% and 66.7%. In addition, a laboratory study was conducted to demonstrate how the system could be used to interact with a dump truck. Future work will integrate the proposed system into a robotic construction machine.
In alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flow charts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and in which sub-operations described as part of larger operations are performed independently.
Furthermore, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise stated to the contrary, one or more of the described functions and/or features may be integrated in a single physical device and/or software module, or one or more functions and/or features may be implemented in a separate physical device or software module. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary for an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be understood within the ordinary skill of an engineer, given the nature, function, and internal relationship of the modules. Accordingly, those skilled in the art can, using ordinary skill, practice the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative of and not intended to limit the scope of the invention, which is defined by the appended claims and their full scope of equivalents.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
While the preferred embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A man-machine interaction system mounted on a construction robot, comprising:
the position tracking module is used for detecting and tracking the position of a construction worker;
the gesture tracking module is used for tracking and identifying hand actions of construction workers when the construction workers are detected to make gestures to the robot;
and the gesture recognition module is used for recognizing the meaning of the gesture of the worker and outputting a corresponding instruction.
2. The human-computer interaction system mounted on a construction robot according to claim 1, wherein the position tracking module comprises:
the method comprises the steps that image information of construction workers is collected in real time through a construction robot, and a first video sequence is established through collected images;
identifying construction workers in the video sequence, and establishing different identification IDs for each construction worker;
drawing a boundary frame of a construction worker, and modeling appearance information of the construction worker;
and drawing the action track of the construction worker through the video sequence and the appearance information of the construction worker, and associating the collected image with the action track to realize the position tracking of the construction worker.
3. The human-computer interaction system mounted on a construction robot as claimed in claim 2, wherein the position tracking module is configured to perform position tracking using a convolutional neural network, and the convolutional neural network comprises a YOLOv3 convolutional neural network.
4. The human-computer interaction system mounted on a construction robot according to claim 2, wherein the operation step of the gesture tracking module comprises:
amplifying the first video sequence with the signaling construction worker as the center when receiving the gesture signal of the construction worker,
adjusting the first video sequence according to the bounding box of the construction worker to obtain a second video sequence;
and performing motion capture on the gestures of the construction workers, and generating and outputting a third video sequence.
5. The human-computer interaction system mounted on a construction robot as claimed in claim 4, wherein the adjusting the first video sequence according to the boundary box of the construction worker specifically comprises: and enabling the distance between the boundary box of the construction worker and the edge of the collected image to be not less than one eighth of the corresponding radial side length of the collected image.
6. The human-computer interaction system mounted on a construction robot according to claim 4, wherein the operation step of the gesture recognition module comprises:
detecting a gesture using a detector;
identifying a specific meaning of a gesture with a classifier when the gesture is detected by the detector;
and outputting an operation instruction corresponding to the specific meaning according to the specific meaning of the gesture.
7. The human-computer interaction system mounted on a construction robot according to claim 6, wherein the gesture recognition module is implemented by a convolutional neural network based on a hierarchical structure.
8. The human-computer interaction system mounted on a construction robot according to claim 6, wherein the detection of the gesture using the detector specifically includes:
cutting the second video sequence into 8 video frames according to unit time;
carrying out frame-by-frame detection on the video frame, and extracting gesture features in the video frame through a ResNet-10 convolutional neural network;
the video frame in which the gesture was detected is marked as the first frame.
9. The human-computer interaction system mounted on a construction robot according to claim 6, wherein the recognition of the specific meaning of the gesture by the classifier specifically comprises:
further cropping the video frames, the second video sequence per unit time being cropped into 32 video frames;
establishing a video frame index T, identifying subsequent video frames from a first frame TO, and classifying gestures in the video frames when the difference between T and T0 is equal TO a multiple of a time factor L;
calculating the weighted probability of each frame in the video frame index T through a weighting function, calculating the difference value between the highest value and the next highest value of the weighted probabilities, and searching a corresponding gesture in a library according to the gesture in the video frame when the difference value is larger than a preset threshold value;
and outputting the specific meaning of the corresponding gesture.
10. The human-computer interaction system mounted on a construction robot as claimed in claim 6, wherein the weighting function is expressed by the formula:
wherein, wTReferring to the weight at the Tth frame, u corresponds to the average duration of the gesture (i.e., the number of frames) in the dataset and s is the stride length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111335163.1A CN114167978A (en) | 2021-11-11 | 2021-11-11 | Human-computer interaction system carried on construction robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111335163.1A CN114167978A (en) | 2021-11-11 | 2021-11-11 | Human-computer interaction system carried on construction robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114167978A true CN114167978A (en) | 2022-03-11 |
Family
ID=80478899
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111335163.1A Pending CN114167978A (en) | 2021-11-11 | 2021-11-11 | Human-computer interaction system carried on construction robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114167978A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426480A (en) * | 2011-11-03 | 2012-04-25 | 康佳集团股份有限公司 | Man-machine interactive system and real-time gesture tracking processing method for same |
| CN104460991A (en) * | 2014-11-19 | 2015-03-25 | 中山大学 | Gesture interaction control system based on digital household equipment |
| CN104992171A (en) * | 2015-08-04 | 2015-10-21 | 易视腾科技有限公司 | Method and system for gesture recognition and man-machine interaction based on 2D video sequence |
| WO2018033154A1 (en) * | 2016-08-19 | 2018-02-22 | 北京市商汤科技开发有限公司 | Gesture control method, device, and electronic apparatus |
| CN108256421A (en) * | 2017-12-05 | 2018-07-06 | 盈盛资讯科技有限公司 | A kind of dynamic gesture sequence real-time identification method, system and device |
| CN111709310A (en) * | 2020-05-26 | 2020-09-25 | 重庆大学 | Gesture tracking and recognition method based on deep learning |
| CN112506342A (en) * | 2020-12-04 | 2021-03-16 | 郑州中业科技股份有限公司 | Man-machine interaction method and system based on dynamic gesture recognition |
| CN113378770A (en) * | 2021-06-28 | 2021-09-10 | 北京百度网讯科技有限公司 | Gesture recognition method, device, equipment, storage medium and program product |
-
2021
- 2021-11-11 CN CN202111335163.1A patent/CN114167978A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102426480A (en) * | 2011-11-03 | 2012-04-25 | 康佳集团股份有限公司 | Man-machine interactive system and real-time gesture tracking processing method for same |
| CN104460991A (en) * | 2014-11-19 | 2015-03-25 | 中山大学 | Gesture interaction control system based on digital household equipment |
| CN104992171A (en) * | 2015-08-04 | 2015-10-21 | 易视腾科技有限公司 | Method and system for gesture recognition and man-machine interaction based on 2D video sequence |
| WO2018033154A1 (en) * | 2016-08-19 | 2018-02-22 | 北京市商汤科技开发有限公司 | Gesture control method, device, and electronic apparatus |
| CN108256421A (en) * | 2017-12-05 | 2018-07-06 | 盈盛资讯科技有限公司 | A kind of dynamic gesture sequence real-time identification method, system and device |
| CN111709310A (en) * | 2020-05-26 | 2020-09-25 | 重庆大学 | Gesture tracking and recognition method based on deep learning |
| CN112506342A (en) * | 2020-12-04 | 2021-03-16 | 郑州中业科技股份有限公司 | Man-machine interaction method and system based on dynamic gesture recognition |
| CN113378770A (en) * | 2021-06-28 | 2021-09-10 | 北京百度网讯科技有限公司 | Gesture recognition method, device, equipment, storage medium and program product |
Non-Patent Citations (1)
| Title |
|---|
| XIN WANG, ZHENHUA ZHU: "《Vision–based framework for automatic interpretation of construction workers’ hand gestures》", 《AUTOMATION IN CONSTRUCTION》, vol. 130, pages 1 - 11 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109657575B (en) | Intelligent video tracking algorithm for outdoor constructors | |
| CN108960067B (en) | Real-time train driver action recognition system and method based on deep learning | |
| Devasena et al. | Video surveillance systems-a survey | |
| CN108596148B (en) | System and method for analyzing labor state of construction worker based on computer vision | |
| CN102103409A (en) | Man-machine interaction method and device based on motion trail identification | |
| Wang et al. | Vision–based framework for automatic interpretation of construction workers' hand gestures | |
| CN104680559A (en) | Multi-view indoor pedestrian tracking method based on movement behavior mode | |
| Rangesh et al. | When vehicles see pedestrians with phones: A multicue framework for recognizing phone-based activities of pedestrians | |
| CN113344967B (en) | Dynamic target identification tracking method under complex background | |
| EP4232340A1 (en) | Rail feature identification system | |
| CN111738218A (en) | Human body abnormal behavior recognition system and method | |
| KR20220078893A (en) | Apparatus and method for recognizing behavior of human in video | |
| US10929688B2 (en) | System and method of video content filtering | |
| CN116259002A (en) | Human body dangerous behavior analysis method based on video | |
| CN109720353B (en) | Driving behavior detection method based on smart phone | |
| CN112785564B (en) | Pedestrian detection tracking system and method based on mechanical arm | |
| CN113537019A (en) | Detection method for identifying wearing of safety helmet of transformer substation personnel based on key points | |
| CN117893998A (en) | Intelligent collision prevention method for pose of human machine based on machine vision | |
| CN109272036A (en) | A kind of random fern method for tracking target based on depth residual error network | |
| Jean et al. | Body tracking in human walk from monocular video sequences | |
| CN110378935B (en) | Parabolic identification method based on image semantic information | |
| CN114167978A (en) | Human-computer interaction system carried on construction robot | |
| Kushwaha et al. | Rule based human activity recognition for surveillance system | |
| Axenbeck et al. | Recognizing complex, parameterized gestures from monocular image sequences | |
| Katuwandeniya et al. | Exact-likelihood User Intention Estimation for Scene-compliant Shared-control Navigation |
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 |