Adaptive Occlusion Detection System and Method in Video Tracking

A technology for occlusion detection and video tracking, which is applied in the field of computer vision and can solve problems such as non-adaptability

Active Publication Date: 2021-07-20
SHANGHAI JIAOTONG UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this occlusion detection method includes too many predefined parameters, and it is not adaptive to different sequences and different targets

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
  • Adaptive Occlusion Detection System and Method in Video Tracking
  • Adaptive Occlusion Detection System and Method in Video Tracking
  • Adaptive Occlusion Detection System and Method in Video Tracking

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] An adaptive occlusion detection system in video tracking, comprising:

[0049] Background tracker, the background tracker uses Kernelized Correlation Filter (Kernelized CorrelationFilter) to track the small blocks in the "Occlusion Candidate Set" (OPC) in the target tracking result of the target tracker, and the weight of the filter can be obtained by the following formula :

[0050] w=argmin(∑(f(w,x i )-y i ) 2 +λ||w|| 2 ) (3)

[0051] where x i is the training sample, y i is the regression target, and λ is the regularization coefficient. via nonlinear mapping Transform the nonlinear relationship between the training samples and the regression target into a linear one, then α can be obtained by the following formula:

[0052] α=(K+λI) -1 Y (4)

[0053] where K is the kernel matrix, Choose an appropriate kernel function k(x i , x j ) can make the kernel matrix K diagonalized by discrete Fourier transform, then the solution of the correlation filter is:...

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 present invention provides a self-adaptive occlusion detection system and method in video tracking, comprising: a background tracker: according to the tracking result of the target by the target tracker, tracking the small background block of the occluded target and the background small blocks around the target; occlusion detection Device: According to the tracking results of the target tracker and the background tracker, judge the positional relationship between the target and the background block, and output the position of the background block that occludes the target; template updater: calculate the degree of occlusion of the target, when When it is less than a threshold, update the target template, and when the occlusion degree is greater than or equal to the threshold, stop updating the target template; search range predictor: change the tracking range of the target tracker to the target according to the degree of occlusion of the target. The present invention introduces a background tracker to track small background blocks, and uses the occlusion information of the previous frame to set an adaptive threshold, so that the relationship between the background and the target can be better judged and the occlusion can be detected more accurately.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to an adaptive occlusion detection system and method in video tracking. Background technique [0002] Video tracking is one of the most important research topics in the field of computer vision, and has important applications in scene monitoring, human-computer interaction, and medical images. Video tracking is to give the initial position of the target in the first frame of the video sequence, and the system predicts the position of the target in the subsequent frames. A video tracking system generally consists of five parts: motion model, feature extractor, observation model, model updater, and ensemble post-processor. Among them, the motion model generates candidate regions that may contain targets based on the estimation of previous frames; the feature extractor performs feature extraction on each candidate region; the observation model judges whether the candidate regi...

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 Patents(China)
IPC IPC(8): G06T7/223
Inventor 乔宇谷月阳
Owner SHANGHAI JIAOTONG 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