Light human body action recognition method based on deep learning

Abstract

The invention discloses a light human body action recognition method based on deep learning. The method comprises the steps of firstly constructing a light-weight deep learning network (SDNet) combining a shallow network and a deep network, wherein the network comprises a shallow multi-scale module and a deep network module, and constructing a light-weight human body action recognition model basedon deep learning based on the network; in the model, firstly utilizing the SDNet to perform feature extraction and representation on the space-time double flow; utilizing a time pyramid pooling layerto aggregate the video frame level features of the time stream and the space stream into the video level representation; obtaining a recognition result of the space-time double flow to the input sequence via a full connection layer and a softmax layer, and finally fusing the double-flow result in a weighted average fusion mode to obtain a final recognition result. By adopting the light human bodyaction recognition method based on the deep learning, the model parameter quantity can be greatly reduced on the premise of ensuring that the recognition precision is not reduced.

Description

technical field [0001] The present invention relates to the technical field of graphics and image processing, in particular to a supervised light-weight human action recognition model method based on deep learning. Background technique [0002] The main problem to be solved in human action recognition is how to analyze and process the video sequences collected by cameras or sensors, so that the computer can "understand" human actions and behaviors in the video, which is important for security monitoring, entertainment methods, etc. The research significance of human action recognition based on video is also widely used in human-computer interaction, virtual reality, smart home equipment and other fields. Human action recognition or understanding of human behavior is essential for many AI systems. For example, the video surveillance system contains hundreds of hours of surveillance video. If you manually traverse the surveillance video, not only is the work tedious and lengt...

AI Technical Summary

Problems solved by technology

[0009] In order to solve the problems of the existing human action recognition model based on deep learning, such as a large number of parameters and too deep and heavy network, the present invention provides a light-weight human action recognition method based on deep learning

Images