Selective Covariance-based Human Localization, Classification and Tracking in Video Streams from Multiple Cameras

Taranyan, A. R.; Devyatkov, Vladimir; Alfimtsev, Alexander

doi:10.5220/0006538100810088

Cited by 3 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Those people whose faces are hidden behind the masking items will be indicated on the security monitor as a person with a heightened threat. At this stage, there are two scenarios: the control over the security of the facility will be fully placed on the shoulders of the security service, and the program will pay attention of the security service to people with difficult-to-recognize persons, after which the guards make a decision on each person [20]; the software independently makes an assumption about each person with a difficult to recognize face, and with the help of a trigger camera, which will be installed at the place of a teller or employee of the hall, monitors the behavior of each employee [5], [14]. This system is able to replace the emergency call button, which will shorten the response time of the security services and increase the degree of security of the object [17].…”

Section: Introductionmentioning

confidence: 99%

Intelligent Video Surveillance Systems with the Purpose of Increasing the Economic Efficiency of Lending Financial Institution

Pisareva¹,

Chekunkov²,

Vakhnin³

2020

Proceedings of the International Scientific Conference "Far East Con" (ISCFEC 2020)

View full text Add to dashboard Cite

The essence of the project is the development of intelligent video surveillance systems and the implementation of sales of hardware and software security systems using face recognition technology. The payback period of the project is ten months. The total revenue for 2020 is planned to be 5, 75 million rubles. The project implementation period is from January 10, 2020, to December 31, 2020. The work calculated the material costs of the company LLC "Point of View" on the development of the project. A new algorithm for recognizing masked people is proposed. The main advantage of the proposed algorithm is that it uses a human mask, so it is applicable for video data with an arbitrary complex background. The results of testing the algorithm on its own sample of video sequences obtained from a stationary video surveillance camera are given.

show abstract

Section: Introductionmentioning

confidence: 99%

Intelligent Video Surveillance Systems with the Purpose of Increasing the Economic Efficiency of Lending Financial Institution

Pisareva¹,

Chekunkov²,

Vakhnin³

2020

Proceedings of the International Scientific Conference "Far East Con" (ISCFEC 2020)

View full text Add to dashboard Cite

show abstract

“…Visual object tracking is a challenging task related to various problem domains including self-driving cars, security, robotics, etc. Broad scientific research in visual object tracking started in 1980s and is still in progress (Anishchenko, 2018;Deviatkov and Lychkov, 2017;Lychkov et al, 2018;Morozov et al, 2015;Taranyan et al, 2018) because real videos include many issues that are difficult to deal with, e. g. moving object occlusion, drastic illumination changes, video noise and blur. Main approaches to visual object tracking are blob-tracking (Wu et al, 2018), optical flow based tracking (Du et al, 2018), neural network tracking (Price et al, 2018), contour-based tracking (Allili and Ziou, 2006), 3D model based tracking (Yu et al, 2018), etc.…”

Section: Introductionmentioning

confidence: 99%

Semi-Automatic Multi-Camera Video Annotation Tool for Object Tracking and Optical Flow Benchmark Creation

Devyatkov¹,

Lychkov²

2019

International Conferences Interfaces and Human Computer Interaction 2019; Game and Entertainment Technologies 2019; And Compute

View full text Add to dashboard Cite

This paper addresses a problem of video annotation for object tracking evaluation. Contemporary object tracking methods relies heavily on optical flow calculation techniques. State of the art optical flow calculation methods are not enough to provide robust and accurate tracking under object occlusion and shading conditions. Thus, comprehensive visual object tracking datasets are required for deep analysis, benchmarking and improvement of object tracking methods. In most cases publicly available real video datasets does not include ground truth for optical flow evaluation, so new datasets might be in demand. Public video annotation tools for object tracking dataset labeling were reviewed. Lack of open source video annotation tools with functions of automatic optical flow markers and object bounding box visual tracking was noticed. In this work we developed a new open video annotation tool that is suitable for object bounding box labeling as well as optical flow annotation in multicamera video recordings. This tool makes it possible to specify the feature point and/or moving object position on the first video frame and track its position automatically in the subsequent video frames. This tool supports manual adjustment of the tracked position if needed. Multi-layer modular architecture of the annotation tool simplifies its testing, maintenance and further extension by the community.

show abstract

Tracking of Moving Objects With Regeneration of Object Feature Points

Lychkov

Alfimtsev

Sakulin

2018

2018 Global Smart Industry Conference (GloSIC)

View full text Add to dashboard Cite

Selective Covariance-based Human Localization, Classification and Tracking in Video Streams from Multiple Cameras

Cited by 3 publications

References 0 publications

Intelligent Video Surveillance Systems with the Purpose of Increasing the Economic Efficiency of Lending Financial Institution

Intelligent Video Surveillance Systems with the Purpose of Increasing the Economic Efficiency of Lending Financial Institution

Semi-Automatic Multi-Camera Video Annotation Tool for Object Tracking and Optical Flow Benchmark Creation

Tracking of Moving Objects With Regeneration of Object Feature Points

Contact Info

Product

Resources

About