|
|
Multimodal System for Multi-Object Tracking in Real-Time
Kučera, Adam ; Šátek, Václav (referee) ; Rozman, Jaroslav (advisor)
This thesis deals with the topic of multi-object multi-sensor tracking. A conventional track-oriented multiple hypothesis tracking (TOMHT) pipeline is implemented in C++ programming language and an implementable interface is designed, enabling to easily extend the core algorithm with arbitrary sensors and measured target attributes, making the system multimodal, i.e.\ applicable in heterogeneous systems of sensors. A novel algorithm for solving combinatorial optimization arising in TOMHT is proposed. Finally, few example implementations of the interface are provided and the system is evaluated in simulated and real-world scenarios.
|
|
|
Moving Object Detection in Video
Buchtela, Radim ; Ševcovic, Jiří (referee) ; Herout, Adam (advisor)
This thesis is devoted to detection of moving object in video sequence. In this thesis are described in detail detection methods based on principle of Background subtraction. There are indtroduced strategies of background modeling and background model updating. This thesis deal also with problems of tracking objects in video, there is introduced one of possible approaches. Part of this thesis is application for function demonstration of detection methods and tracking method. On results are showed main advantages and disadvantages of implemented methods.
|
|
|
Ball Tracking in Sports Video
Motlík, Matúš ; Špaňhel, Jakub (referee) ; Bartl, Vojtěch (advisor)
This master's thesis deals with automatic detection and tracking of a soccer ball in sports videos. Based on the introduced techniques focusing on tracking of small objects in high-resolution videos, effective convolutional neural networks are designed and used by a modified version of tracking algorithm SORT for automatic object detection. A set of experiments with the processing of images in different resolutions and with various frequencies of detection extraction is carried out in order to examine the trade-off between processing speed and tracking accuracy. The obtained results of experiments are presented and used to form proposals for future work, which could lead to improvements in tracking accuracy while maintaining reasonable processing speed.
|
|
|
Counting People on the FIT Foot Bridge
Maliňák, Václav ; Nečas, Ondřej (referee) ; Herout, Adam (advisor)
This thesis is devoted to determining the number of moving object in video sequences. It explains principles and basic methods of moving object detection in video sequences. Methods such as LBP, optical flow and intensity histograms are mentioned, background subtraction method is examined more deeply. Afterwards, the principles of object tracking is explained, as well as possible methods of object counting. An important part of this thesis constitutes implementation of an application that was tested on videos capturing people at move over the FIT foot bridge. At the end of this thesis, application measure of success is evaluated.
|
|
|
Web Application for Inspecting Results of Automatic Video Processing and Manual Annotations
Červíček, Petr ; Švec, Tomáš (referee) ; Smrž, Pavel (advisor)
The thesis pursues the implementation of the web application for obtaining valuable data for anotation. Data are mainly collected from videos and images, but they can be obtained from deepfakes images and gifs as well. The YOLO - You Only Look Once is used for object detection in the videos. Whole application is developed in node.js and react. Node.js for backend and react for frontend. For a better understanding of web application, there is also historical description of the web technologies.
|
|
|
Object Tracking in a Sequence of Images
Švancar, Boris ; Doležel, Michal (referee) ; Orság, Filip (advisor)
The bachelor thesis deals with motion detection and object tracking in video sequence from static and dynamic cameras. The background subtraction method along with using of background updating and Lucas-Kanade optical flow method are introduced in my thesis as well. Both methods are implemented in the test application in the language C++ using OpenCV and Qt libraries and consequently both of them are tested. The results are evaluated in conclusion of the bachelor thesis.
|
|
|
Analysis of Surveillance Camera Recordings
Ščavnická, Šárka ; Švec, Tomáš (referee) ; Smrž, Pavel (advisor)
This thesis deals with the systems for analyzing records from security cameras. It aims to create a functional solution that analyzes records and answers questions from the user. The created system combines the YOLO algorithm for object detection and DeepSORT for their subsequent tracking. It contains five models that detect specific situations. Individual models achieved varying degrees of success during testing, with the lowest success rate being 58 % for the getting out of car situation. The highest success rate, 83 %, was obtained by a model for detecting a meeting between two people.
|
|
|
Multi-Person Tracking in Video from Mono-Camera
Vojvoda, Jakub ; Španěl, Michal (referee) ; Beran, Vítězslav (advisor)
Multiple person detection and tracking is challenging problem with high application potential. The difficulty of the problem is caused mainly by complexity of scene and large variations in articulation and appearance of person. The aim of this work is to design and implement system capable of detecting and tracking people in video from static mono-camera. For this purpose, an online method for tracking has been proposed based on tracking-by-detection approach. The method combines detection, tracking and fusion of responses to achieve accurate results. The implementation was evaluated on available dataset and the results show that it is suitable to use for this task. A method for motion segmentation was proposed and implemented to improve the tracking results. Furthermore, implementation of detector based on histogram of oriented gradients was accelerated by taking advantage of graphics processing unit (GPU).
|
|
|
Gestural Interface for Computer Control
Burdík, Vojtěch ; Beneš, Radek (referee) ; Přinosil, Jiří (advisor)
This work focuses on the relatively new computer branch – a computer vision. Work is concentrated on object recognition, localization and reaction to specific movements. The aim is to build a program that will be able to use hand movements to control computer, to the gesture as defined by the fingers and palms respond and make it the event. The objective should be achieved without the use of special sensors or more cameras, but only with one common webcam. For the image processing function from OpenCV library were used and program was built. Evaluation of function is based on functional verification in practice, evaluation of speed, accuracy and quality of object recognition. Following the scheme, how could program for gestural controlling the computer should look, was created, and then his realization in a simple program was made. This program is able to recognize the object in an image, in this case a hand, and through an evaluation of its movements control cursor on a computer with a possibility of simulation pressing the left mouse button. Program continuously allows the documents to scroll horizontal and vertical direction. The program is very robust and work with him is very simple.
|
|
|
Detection and Tracking of Small Moving Objects
Záděra, David ; Mlích, Jozef (referee) ; Juránek, Roman (advisor)
This thesis summarizes and describes the methods for detecting and tracking moving objects in video. The detection methods discussed in the working are background subtraction, segmentation and detection methods using classifiers. From tracking methods are presented Kalman filter, TLD method and particle filter, which is described in detail in a separate chapter. Particle filter is a tracking method that calculates a new estimate of the state of the object based on a set of particles. These particles are defined by their positions and weights. For each next state are these values iteratively updated by measurment and resampling. From the acquired knowledge and expertise has been created a program that demonstrates the activity of the particle filter. This program is part of the DVD, where are also the samples and test results.
|
guest ::
