Faster-RCNN-based gesture recognition system and method
A gesture recognition and gesture technology, applied in the field of gesture recognition system based on Faster-RCNN, can solve the problem of large consumption of computing resources, achieve the effect of improving efficiency, reducing computing performance requirements, improving accuracy and
Pending Publication Date: 2021-07-30
NORTHEASTERN UNIV LIAONING
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a gesture recognition system and method based on Faster-RCNN, which can realize high-accuracy gesture recognition only by using a monocular camera to capture user gestures, so that users can Interact with the virtual scene through gestures, and at the same time solve the problem of large consumption of computing resources on a single computer
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0037] The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0038] Since the user interacts with the virtual scene through gestures, the performance requirements of a single computer are relatively high, so the gesture recognition method and system based on Faster-RCNN (Faster Region Based Convolutional Neural Networks, Agile Region Convolutional Neural Networks) provided by the present invention In, including the client and server, in order to reduce the performance requirements of a single computer. Hand gesture images are processed on the client side, and hand images are preprocessed using bilateral filtering and moving target tracking methods based on background subtraction, capturing hand areas and removing complex backgrounds...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a Faster-RCNN-based gesture recognition system and method, and belongs to the technical field of computer vision. The system comprises: a monocular camera used for collecting a real-time gesture image and sending the real-time gesture image to a client; the client, which is used for processing each gesture image to capture a hand contour, accumulating a certain number of gesture images which are processed and marked with gesture categories to form a gesture image data set, and sending the gesture image data set to a server, and, after the gesture image data set is obtained, sending the processed real-time gesture image to the server; the server, which is used for training the Faster-RCNN by using the gesture image data set to obtain a gesture recognition model, and recognizing the real-time gesture of the user from the received real-time gesture image through the trained gesture recognition model. High-accuracy gesture recognition can be realized only by using the monocular camera to collect user gestures, and meanwhile, the problem of high computing resource consumption of a single computer is solved.
Description
technical field [0001] The invention relates to the technical field of computer vision, in particular to a gesture recognition system and method based on Faster-RCNN. Background technique [0002] Gesture recognition technology is an important method for users to interact with virtual reality scenes. Gestures can be defined as hand gestures, which can be tracked by a computer to obtain the information contained in them and convert them into meaningful instructions. Gesture recognition technology is often used in virtual reality applications based on somatosensory interaction. When a user interacts with a virtual reality scene, the computer needs to obtain and recognize the user's gestures, and synchronize the pose information of the user's hand to the virtual reality scene. Objects in interact. [0003] Most of the existing gesture recognition methods use data gloves, depth cameras or monocular cameras to acquire user gestures. Data gloves are devices that collect gesture ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06T7/194
CPCG06T7/194G06V40/117G06V40/113
Inventor 高天寒杨镇豪
Owner NORTHEASTERN UNIV LIAONING