Figure video clip extraction method based on multi-target motion tracking

A technology for video clips and motion tracking, applied in closed-circuit television systems, image analysis, image enhancement, etc., which can solve the impact of accurate recognition, face recognition methods that are not applicable to video images, and uneven data quality of face recognition sample sets, etc. problem, to achieve less time-consuming effect

Active Publication Date: 2019-07-12
WUHAN UNIV
View PDF7 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] 1. The face detection process takes a long time, which is not conducive to real-time analysis and large-scale processing of video streams;
[0005] 2. Video images with no face part or part of the face covered are not suitable for the face recognition method;
[0006] 3. The data quality of the face recognition sample set is uneven, which has a negative impact on accurate recognition

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
  • Figure video clip extraction method based on multi-target motion tracking
  • Figure video clip extraction method based on multi-target motion tracking
  • Figure video clip extraction method based on multi-target motion tracking

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0039] please see figure 1 , the technical solution adopted by the present invention is: a method for extracting video clips of people based on multi-target motion tracking, which first extracts video clips where moving objects appear, and then performs identification of people on the extracted video clips. The main steps are: use the frame difference method to detect the moving target in the video frame image to be tested; track the motion trajectory of the detected moving target; frame the tracked moving target; extract the video according to the start frame and end frame where the moving target appears Segments; Person identification is performed on extracted video segments.

[0040] please see figure 2 , in this specific embodiment, the moving object detection is performed from the third frame of the video to be tested, assuming that the...

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 provides a figure video clip extraction method based on multi-target motion tracking, which comprises the following steps: firstly, extracting a video clip in which a moving target appears, and then carrying out figure identity recognition on the extracted video clip. The method is used for monitoring a video shot by a camera, that is, the camera is fixed within a long period of time, a video with no large change in the background of a picture is shot, all moving targets in the video picture can be tracked in real time, the moving targets are identified and distinguished, a videosegment in which a moving object appears is extracted, and the identity of a person in the moving object is identified.

Description

technical field [0001] The invention belongs to the technical field of video image processing, in particular to a method for extracting character video segments based on multi-target motion tracking. Background technique [0002] Multi-target motion tracking is to simultaneously locate and track multiple moving targets in the video to be detected, maintain the trajectory of the target movement, and identify and distinguish different targets. At present, the realization of multi-target motion tracking requires some higher-level methods to complete, such as: pose estimation, action recognition, behavior analysis, etc. [0003] Most of the video clip extraction methods for person identity are based on technologies and processes such as face detection and face recognition. These techniques and processes create the following major issues in practical application: [0004] 1. The face detection process takes a long time, which is not conducive to real-time analysis and large-sca...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/215H04N7/18
CPCG06T2207/30201G06T7/215H04N7/18
Inventor 赵鹏翔
Owner WUHAN UNIV
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap