R-C3D network-based end-to-end video time sequence behavior detection method

Abstract

The invention discloses an R-C3D network-based end-to-end video time sequence behavior detection method, which belongs to the field of computer vision, and comprises the steps of performing frame rateadjustment and frame extraction on an input video, and performing normalization and data enhancement on extracted frames to obtain a training set and a test set; constructing a time sequence behaviordetection model, wherein the time sequence behavior detection model comprises a feature extraction module, a long-time information coding module and a behavior identification module, the long-time information coding module is used for coding an extracted feature map to obtain features containing long-time information; inputting the training set and the test set into a time sequence behavior detection model for training; and inputting a to-be-detected video into the trained time sequence behavior detection model for detection to obtain a behavior category and positioning information existing in the video. According to the method, the extracted features are encoded by designing the long-term information encoding network, so that the network can obtain global time information of time sequence actions, and the accuracy of action positioning and classification is improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and more specifically relates to an end-to-end video timing behavior detection method based on an R-C3D network. Background technique [0002] With the popularity of video, mobile phones, and the Internet, more and more videos are produced. In order to be able to search videos in a targeted manner, and find specific behaviors from videos, such as stealing, fighting and other dangerous behaviors, so as to ensure property and personal safety. Temporal action detection in long untrimmed videos has received increasing attention. The goal of this task is to temporally localize intervals of action occurrences of interest and predict action categories in long untrimmed videos. Action detection tasks face many challenges due to the large variation in the temporal duration of actions, from seconds to fractions in between, and the fact that the target behavior is in the context of unrelated activ...

AI Technical Summary

Problems solved by technology

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides an end-to-end video timing behavior detection method based on the R-C3D network, the purpose of which is to solve the problem of low detection accuracy and low operating speed of the existing timing behavior detection method. slow technical issues

Images