Comparison of tracking algorithms implemented in OpenCV

Janku, Peter; Koplík, Karel; Dulík, Tomáš; Szabó, István

doi:10.1051/matecconf/20167604031

Cited by 23 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since MedianFlow tracker [5] and MIL tracker [6] are not adapted to our application (random displacement, quick rotation), we will focus on the other trackers. For a more exhaustive comparison, please see the work of Janku et al [8].…”

Section: Available Techniquesmentioning

confidence: 99%

Automatic Video Editing: Original Tracking Method Applied to Basketball Players in Video Sequences

Nost¹,

Lefèvre²,

Bombardier³

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The main task here is to track several basketball players during a game and to be able to retrieve their whole trajectories at the end. The final application is to get some statistics about each players and to identify some special events like free throw or to determine when a counterattack is going to happen. The originality of the solution states in the way the tracking is performed : instead of studying the close environment of each player, all the players are detected on each frame then we are using specific informations like background, speed vector, color or distance between players to link player's positions and create the whole trajectories. We will compare our results with a benchmark of algorithms to see that our solution is quite efficient in term of tracking and speed.

show abstract

Section: Available Techniquesmentioning

confidence: 99%

Automatic Video Editing: Original Tracking Method Applied to Basketball Players in Video Sequences

Nost¹,

Lefèvre²,

Bombardier³

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…position and size) of the object in the initial frame. Closely related are feature detectors; those will just detect a specific object on each individual frame, based purely on matching features [3]. Pure trackers, like investigated in this study, aim to follow the trajectory as well as predict the future location of the object.…”

Section: Introductionmentioning

confidence: 99%

New Object Tracker Based On Adaptive Intensity Models of Object and Its Surroundings

Gors¹,

Hofman²,

Birem³

et al. 2021

World Congress on Electrical Engineering and Computer Systems and Science

View full text Add to dashboard Cite

New developments on the object tracker topic are needed, so that reliable tracking systems can have value for industrial applications, like surveillance and assembly monitoring. This paper presents a new object tracker algorithm based on adaptive models of the intensity probabilities of the object and its surroundings. Using the tracked object contour in the previous frame and the object path allow to estimate a narrow search area, in which contours with high object probability are combined, after masking pixels with high surrounding probabilities away. Rules about the object contour area ensures that the tracked contour doesn't drift away between frames or spreads into the surroundings. If the tracking is lost, the contour prediction in combination with the surrounding estimation takes over, filling the gaps until the object intensity-based tracker leads the tracking again. The proposed tracker was contrasted against three of the available trackers in OpenCV (i.e. KCF, CRST and MOSSE). Their performances were evaluated on two different applications (i.e. drone tracking and part tracking in an assembly cell) based on the Intersection over Union (IoU)-metric and their processing time. The obtained results show that the proposed tracker is faster and more accurate.

show abstract

“…In addition, if the best algorithm fails on speci c frames, the possibility of re-tracking this frame using another tracking algorithm was checked out [31].…”

mentioning

confidence: 99%

The comparison of OpenCV library algorithms for tracking movements of liver tumors in ultrasound video.

Levin

Klimov

Nechunaev

et al. 2022

Preprint

View full text Add to dashboard Cite

This study aims to compare the tracking algorithms provided by the OpenCV library to use on ultrasound video. Despite the widespread application of this computer vision library, few works describe the attempts to use it to track the movement of liver tumors on ultrasound video. Movements of the neoplasms caused by the patient`s breath interfere with the positioning of the instruments during the process of biopsy and radio-frequency ablation. The main hypothesis of the experiment was that tracking neoplasms and correcting the position of the manipulator in case of using robotic-assisted surgery will allow positioning the instruments more precisely. Another goal of the experiment was to check if it is possible to ensure real-time tracking with at least 25 processed frames per second. OpenCV version 4.5.0 was used with 7 tracking algorithms from the extra modules package. They are: Boosting, CSRT, KCF, MedianFlow, MIL, MOSSE, TLD. More than 4900 frames of standard definition were processed during the experiment. Analysis of the results shows that two algorithms – CSRT and KCF – could solve the problem of tumor tracking. They lead the test with more than 70% of ROI coverage and more than 95% successful searches. They could also be used in real-time processing with an average processing speed of up to 30 frames per second in CSRT and up to 100 frames per second for KCF. This experiment also shows that no frames made CSRT and KCF algorithms fail simultaneously. So, the hypothesis for future work is combining these algorithms to work together, with one of them – CSRT – as support for the KCF tracker on the rarely failed frames.

show abstract

Comparison of tracking algorithms implemented in OpenCV

Cited by 23 publications

References 12 publications

Automatic Video Editing: Original Tracking Method Applied to Basketball Players in Video Sequences

Automatic Video Editing: Original Tracking Method Applied to Basketball Players in Video Sequences

New Object Tracker Based On Adaptive Intensity Models of Object and Its Surroundings

The comparison of OpenCV library algorithms for tracking movements of liver tumors in ultrasound video.

Contact Info

Product

Resources

About