Video behavior recognition method based on weighted fusion of multiple image tasks

A technology of weighted fusion and recognition method, which is applied in the direction of character and pattern recognition, instruments, computer parts, etc., can solve the problems of time-consuming and labor-intensive, and achieve the effect of expanding and reducing the distance

Pending Publication Date: 2021-10-22
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] The present invention fully considers the correlation between different visual tasks in the field of computer vision, and the characteristics that the prior knowledge of related tasks can be transferred and utilized, and proposes a video behavior rec

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Video behavior recognition method based on weighted fusion of multiple image tasks
  • Video behavior recognition method based on weighted fusion of multiple image tasks
  • Video behavior recognition method based on weighted fusion of multiple image tasks

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0051] Example

[0052] Step 1, collect the human activity video data set, divide according to the human body behavior category in the video, and give the category tag, the video data is subjected to normalization, divided into training set and test set, the specific method is:

[0053] Step 1.1. The collection of video data includes the self-built video data set or use the existing public data set: first download the relevant data set file from the official website, the specific data set is: HMDB51 is a video behavior identification with 51 action tags. There are 6849 videos, and each action contains at least 51 videos. The action mainly includes: facial action such as smiling, chewing, talking, facial and goods interacting, eating, drinking; body movements such as clapping, climbing, jumping, Run, interact with the item, such as combing, driver, playing golf, people with interaction between people, like hug, kiss; each type of action is doing action by 25 people, 1,3320 videos I...

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

No PUM Login to view more

Abstract

The invention relates to a video behavior recognition method based on weighted fusion of multiple image tasks. The video behavior recognition method comprises the following specific steps: 1, constructing an initialized teacher network; 2, downloading and selecting a plurality of pre-training models and parameters of a visual image task common data set in positive correlation with video behavior identification as an initialized teacher network; 3, establishing a multi-teacher video behavior recognition knowledge base; 4, under the guidance of the multi-teacher network with the weight distributed again, carrying out self-supervised training based on comparative learning on the student network; 5, carrying out performance test on the model video behavior identification on the test data set. The method has the advantages that the image task in positive correlation with the video behavior recognition task serving as the target task is used as the teacher task, the training mode of comparison self-supervised learning is adopted, and the video behavior recognition problem under the condition that high-quality video marking samples are insufficient is solved. Therefore, the accuracy of video behavior identification is effectively improved.

Description

technical field [0001] The invention relates to the technical field of video behavior analysis, in particular to the design of a video behavior recognition method for weighted fusion of multiple image tasks. Background technique [0002] Behavior recognition is an attractive and challenging research direction in recent years, that is, given a cropped video, judge the human behavior category in this video through computer vision technology. The development of deep convolutional neural networks and the emergence of large-scale labeled datasets in recent years have significantly improved the accuracy of action recognition. Behavior recognition technology is playing an increasingly important role in many fields such as intelligent security, human-computer interaction, video understanding, and medical health. [0003] The existing supervised learning-based Deep Convolutional Neural Network (Deep CNN) model algorithm has achieved relatively ideal results. However, in order to ob...

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
no application Login to view more
IPC IPC(8): G06K9/00
CPCG06F18/22G06F18/2155G06F18/2415G06F18/253
Inventor 高广宇刘驰李金洋
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products