Human body behavior recognition method and system based on human body skeleton
A human skeleton and recognition method technology, applied in character and pattern recognition, neural architecture, instruments, etc., can solve the problems that the graph network cannot be iterated, difficult to generate connections, and lose spatial information, etc., to reduce training parameters and training costs, and improve Effects of applying capabilities, reducing computation time
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0036] like figure 1 As shown, the present embodiment provides a human body behavior recognition method based on a human skeleton, including:
[0037] S1: Obtain the behaviors and actions of the human skeleton and the corresponding skeleton point coordinates, coordinate differences between frames of skeleton points and skeleton features, and construct a training set;
[0038] S2: According to the training set, the graph convolutional network and the attention mechanism network based on human body parts are trained sequentially, and the behavior recognition model is constructed with the trained graph convolutional network and attention mechanism network;
[0039] S3: Recognize the human skeleton to be recognized according to the behavior recognition model, and output human behavior actions.
[0040] In the step S1, a skeleton sequence is regarded as a graph, and the skeleton points in each frame are connected according to the natural human body structure, and there are also co...
Embodiment 2
[0081] This embodiment provides a human body behavior recognition system based on a human skeleton, including:
[0082] The data acquisition module is used to obtain the behavior of the human skeleton and the corresponding skeleton point coordinates, the coordinate difference between the skeleton point frames and the skeleton features, and construct the training set;
[0083] The training module is used to sequentially train the graph convolution network and the attention mechanism network based on human body parts according to the training set, and construct the behavior recognition model with the trained graph convolution network and attention mechanism network;
[0084] The identification module is configured to identify the skeleton of the human body to be identified according to the behavior identification model, and output human behavior actions.
[0085] It should be noted here that the above-mentioned modules correspond to steps S1 to S3 in Embodiment 1, and the exampl...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com