SciTePress - Affiliation Profile

Hasselt University, Belgium

Context (please use complete words only)

Field

Sort by

48 Papers

Real-time Detection of 2D Tool Landmarks with Synthetic Training Data

Topics: Image and Video Processing; Deep Learning; Scene Analysis and Understanding

Authors: Bram Vanherle , Jeroen Put , Nick Michiels and Frank Van Reeth

Affiliation: Hasselt University - tUL - Flanders Make, Expertise Centre for Digital Media, Agoralaan, 3590 Diepenbeek, Belgium

Digital Object Identifier: 10.5220/0010689900003061

Proceedings of ROBOVIS 2021, ISBN: 978-989-758-537-1

Quick Abstract

In this paper a deep learning architecture is presented that can, in real time, detect the 2D locations of certain landmarks of physical tools, such as a hammer or screwdriver. To avoid the labor of manual labeling, the network is trained on synthetically generated data. Training computer vision models on computer generated images, while still achieving good accuracy on real images, is a challenge due to the difference in domain. The proposed method uses an advanced rendering method in combination with transfer learning and an intermediate supervision architecture to address this problem. It is shown that the model presented in this paper, named Intermediate Heatmap Model (IHM), generalizes to real images when trained on synthetic data. To avoid the need for an exact textured 3D model of the tool in question, it is shown that the model will generalize to an unseen tool when trained on a set of different 3D models of the same type of tool. IHM is compared to two existing approaches to keypoint detection and it is shown that it outperforms those at detecting tool landmarks, trained on synthetic data.

A Fuzzy Cognitive Map Approach to Investigate the Sustainability of the Social Security System in Jordan

Topics: Applications of Expert Systems; Advanced Applications of Fuzzy Logic; Data Mining; Problem Solving

Authors: George Sammour ¹ , Ahmad Alghzawi ² and Koen Vanhoof ²

Affiliations: ¹ Business Technology Department, Princess Sumaya University for Technology, Amman, Jordan , ² Department of Business Informatics, Hasselt University, Hasselt, Belgium

Digital Object Identifier: 10.5220/0009128304810489

Proceedings of ICEIS 2020, ISBN: 978-989-758-423-7

Quick Abstract

Fuzzy Cognitive Maps are emerging as an important new tool in economic modelling. The aim of this study is to investigates the use of fuzzy cognitive maps with their learning algorithms, based on genetic algorithms, for the purposes of prediction of economic sustainability. A Case study data are extracted from the Jordanian Social Security system for the last 120 months; The Real-Code genetic algorithm and structure optimization algorithm were chosen for their ability to select the most significant relationships between the concepts and to predict future development of the Jordanian social security revenues and expenses. The study shows that fuzzy cognitive maps models clearly predict the future of a complex financial system with incoming and outgoing flows. Therefore, this research confirms the benefits of fuzzy cognitive maps applications as a tool for scholarly researchers, economists and policy makers.

Of the Utmost Importance: Resource Prioritization in HTTP/3 over QUIC

Topics: Web Tools and Languages; New Trends in Internet Technology; Application, Research project and Internet Technology; Technical Infrastructures Suporting Web Applications ; Protocols, Standards, Interoperability and Mobility

Authors: Robin Marx ¹ , Tom De Decker ¹ , Peter Quax ² and Wim Lamotte ¹

Affiliations: ¹ Hasselt University – tUL – EDM, Diepenbeek and Belgium , ² Hasselt University – tUL – EDM, Diepenbeek, Belgium, Flanders Make and Belgium

Digital Object Identifier: 10.5220/0008191701300143

Proceedings of WEBIST 2019, ISBN: 978-989-758-386-5

Quick Abstract

Not even five years after the standardization of HTTP/2, work is already well underway on HTTP/3. This latest version is necessary to make optimal use of that other new and exiting protocol: QUIC. However, some of QUIC’s unique characteristics make it challenging to keep HTTP/3’s functionalities on par with those of HTTP/2. Especially the efforts on adapting the prioritization system, which governs how multiple resources can be multiplexed on a single QUIC connection, have led to some difficult to answer questions. This paper aims to help answer some of those questions by being the first to provide experimental evaluations and result comparisons for 11 different possible HTTP/3 prioritization approaches in a variety of simulation settings. We present some non-trivial insights, discuss advantages and disadvantages of various approaches, and provide results-backed actionable advice to the standardization working group. We also help foster further experimentation by contributing our complete HTTP/3 implementation, results dataset and custom visualizations to the community.

A Multi-Channel Software Decoder for the LoRa Modulation Scheme

Topics: Internet of Things; Software Architecture and Middleware ; Networking and Communication Protocols ; Machine to Machine Communications ; Next Generation Networks; Wireless Systems and Applications; Communication

Authors: Pieter Robyns ¹ , Peter Quax ² , Wim Lamotte ¹ and William Thenaers ¹

Affiliations: ¹ Hasselt University - tUL, Belgium , ² Hasselt University - tUL - Flanders Make, Belgium

Digital Object Identifier: 10.5220/0006668400410051

Proceedings of IoTBDS 2018, ISBN: 978-989-758-296-7

Quick Abstract

LoRa is a recently introduced modulation scheme specifically designed for Low-Power Wide-Area Networks. In this paper, we provide the first detailed and complete description of the LoRa PHY layer, and present a novel methodology for detecting and decoding LoRa frames using Software Defined Radios. Our proposed decoding approach can efficiently decode multiple channels simultaneously in software due to an invariance towards the signal frequency. Hence, our approach also removes the need for correcting frequency offset errors imparted by the transmitter or receiver. We have evaluated our decoding technique in a lab setup using three Software Defined Radios (USRP B210, HackRF, and RTL-SDR) and three commercial off-the-shelf hardware LoRa transceivers (Microchip RN2483, HopeRF RFM96, and Semtech SX1272). We show that our decoder is fully compatible with all configurations of the RN2483 and SX1272, achieving an overall packet error rate of 0 for a signal-to-noise ratio of 20 dB. The source code of the decoder and datasets used in the evaluation are made available publicly.

Related Ontology Subjects/Areas/Topics (7)

Data Communication Networking	Sensor Networks	Software and Architectures
Enterprise Information Systems	Software Agents and Internet Computing	Telecommunications
Internet of Things

IUPTIS: A Practical, Cache-resistant Fingerprinting Technique for Dynamic Webpages

Topics: Web Information Filtering and Retrieval; Web Security and Privacy, Cyber Criminality and Internet, Dark Web

Authors: Mariano Di Martino ¹ , Pieter Robyns ¹ , Peter Quax ² and Wim Lamotte ¹

Affiliations: ¹ Hasselt University/tUL - Expertise Center for Digital Media and Belgium , ² Hasselt University/tUL/Flanders Make - Expertise Center for Digital Media and Belgium

Digital Object Identifier: 10.5220/0007226501020112

Proceedings of WEBIST 2018, ISBN: 978-989-758-324-7

Quick Abstract

Webpage fingerprinting allows an adversary to infer the webpages visited by an end user over an encrypted channel by means of network traffic analysis. If such techniques are applied to websites that contain user profiles (e.g. booking platforms), they can be used for personal identification and pose a clear privacy threat. In this paper, a novel HTTPS webpage fingerprinting method - IUPTIS - is presented, which accomplishes precisely this, through identification and analysis of unique image sequences. It improves upon previous work by being able to fingerprint webpages containing dynamic rather than just static content, making it applicable to e.g. social network pages as well. At the same time, it is not hindered by the presence of caching and does not require knowledge of the specific browser being used. Several accuracy-increasing parameters are integrated that can be tuned according to the specifics of the adversary model and targeted online platform. To quantify the real-world applicability of the IUPTIS method, experiments have been conducted on two popular online platforms. Favorable results were achieved, with a F1 scores between of 82% and 98%, depending on the parameters used. This makes the method practically viable as a means for personal identification.

Concatenation, Embedding and Sharding: Do HTTP/1 Performance Best Practices Make Sense in HTTP/2?

Topics: Protocols and Standards; Metrics and Performance; Web Tools and Languages; New Trends in Internet Technology

Authors: Robin Marx , Peter Quax , Axel Faes and Wim Lamotte

Affiliation: UHasselt-tUL-imec, Belgium

Digital Object Identifier: 10.5220/0006364101600173

Proceedings of WEBIST 2017, ISBN: 978-989-758-246-2

Quick Abstract

Web page performance is becoming increasingly important for end users but also more difficult to provide by web developers, in part because of the limitations of the legacy HTTP/1 protocol. The new HTTP/2 protocol was designed with performance in mind, but existing work comparing its improvements to HTTP/1 often shows contradictory results. It is unclear for developers how to profit from HTTP/2 and whether current HTTP/1 best practices such as resource concatenation, resource embedding, and hostname sharding should still be used. In this work we introduce the Speeder framework, which uses established tools and software to easily and reproducibly test various setup permutations. We compare and discuss results over many parameters (e.g., network conditions, browsers, metrics), both from synthetic and realistic test cases. We find that in most non-extreme cases HTTP/2 is on a par with HTTP/1 and that most HTTP/1 best practices are applicable to HTTP/2. We show that situations in which HTTP/2 currently underperforms are mainly caused by inefficiencies in implementations, not due to shortcomings in the protocol itself.

Related Ontology Subjects/Areas/Topics (3)

Internet Technology

Protocols and Standards

Web Information Systems and Technologies

Simplification of Moving 3D Scene Data on GPU

Topics: Multimedia Databases, Indexing, Recognition and Retrieval; Image and Video Processing, Compression and Segmentation

Authors: Rajesh Chenchu , Nick Michiels , Sammy Rogmans and Philippe Bekaert

Affiliation: Hasselt University, Belgium

Digital Object Identifier: 10.5220/0005962000950098

Proceedings of SIGMAP 2016/ICETE, ISBN: 978-989-758-196-0

Quick Abstract

Real-time large scale continuous image and geometry based data visualization, with an uninterrupted content delivery, quality and rendering, on home and mobile devices is difficult or even mostly impossible because of the low processing capabilities of these hardware devices. However, a gracefully simplified version of the same data can enable us to view the content without significant quality degradation. To do this in a graceful manner, we extended a well-known concept - called ’billboard cloud’ - for animated scene data and implemented this technique using the capabilities of the GPU to generate the simplified versions of large scale data sets.

Related Ontology Subjects/Areas/Topics (6)

Image and Video Processing, Compression and Segmentation		Multimedia Databases, Indexing, Recognition and Retrieval		Multimedia Systems and Applications
Multimedia		Multimedia Signal Processing		Telecommunications

Computing Corpus Callosum as Biomarker for Degenerative Disorders

Topics: Image Registration; Segmentation and Grouping; Medical Image Applications

Authors: Thomas Kovac , Sammy Rogmans and Frank Van Reeth

Affiliation: Hasselt University - tUL - iMinds, Belgium

Digital Object Identifier: 10.5220/0005310201380149

Proceedings of VISAPP 2015/VISIGRAPP, ISBN: 978-989-758-091-8

Quick Abstract

The developed framework can automatically extract a plane with minimal corpus callosum area while simultaneously segmenting it. The method used, introduced by Ishaq, treats the corpus callosum area as a function of the plane extraction parameters and it uses deformable registration to generate a displacement field that can be used for the calculation of the corpus callosum area. Our registration framework is accelerated using CUDA, which enables researchers to benchmark huge amounts of data (patients) to test the hypothesis of the corpus callosum evolution as a biomarker for multiple degenerative disorders like e.g. Alzheimer disease and multiple sclerosis (MS).

Related Ontology Subjects/Areas/Topics (6)

Applications and Services		Image and Video Analysis		Medical Image Applications
Computer Vision, Visualization and Computer Graphics		Image Registration		Segmentation and Grouping

Increased Charge Storage Capacity of Titanium Nitride Electrodes by Deposition of Boron-doped Nanocrystalline Diamond Films

Topics: Neuro-Interface Prosthetic Devices; Electrical Stimulation; Functional Electrical Stimulation (FES)

Authors: Suzan Meijs ¹ , Matthew McDonald ² , Søren Sørensen ³ , Kristian Rechendorff ³ , Vaclav Petrak ⁴ , Milos Nesladek ² , Nico Rijkhoff ¹ and Cristian Pablo Pennisi ¹

Affiliations: ¹ Aalborg University, Denmark , ² University of Hasselt, Belgium , ³ Danish Technological Institute, Denmark , ⁴ Nano6, Czech Republic

Digital Object Identifier: 10.5220/0005606401060109

Proceedings of NEUROTECHNIX 2015, ISBN: 978-989-758-161-8

Quick Abstract

The aim of this study was to investigate the feasibility of depositing a thin layer of boron-doped nanocrystalline diamond (B-NCD) on titanium nitride (TiN) coated electrodes and the effect this has on charge injection properties. The charge storage capacity increased by applying the B-NCD film, due to the wide potential window typical for B-NCD. The impedance magnitude was higher and the pulsing capacitance lower for B-NCD compared to TiN. Due to the wide potential window, however, a higher amount of charge can be injected without reaching unsafe potentials with the B-NCD coating. The production parameters for TiN and B-NCD are critical, as they influence the pore resistance and thereby the surface area available for pulsing.

Related Ontology Subjects/Areas/Topics (6)

Electrical Stimulation		Neuro-Interface Prosthetic Devices		NeuroSensing and Diagnosis
Health Engineering and Technology Applications		Neuromodulation and Neural Prosthesis		Neurotechnology, Electronics and Informatics

Web-mediated Augmentation and Interactivity Enhancement of Omni-directional Video in Both 2D and 3D

Topics: System Integration; Web Services and Web Engineering; Multimedia and User Interfaces; Protocols and Standards; Web Programming

Authors: Maarten Wijnants , Kris Van Erum , Peter Quax and Wim Lamotte

Affiliation: Hasselt University, Belgium

Digital Object Identifier: 10.5220/0005442900210034

Proceedings of WEBIST 2015, ISBN: 978-989-758-106-9

Quick Abstract

Video consumption has since the emergence of the medium largely been a passive affair. This paper proposes augmented Omni-Directional Video (ODV) as a novel format to engage viewers and to open up new ways of interacting with video content. Augmented ODV blends two important contemporary technologies: Augmented Video Viewing and 360 degree video. The former allows for the addition of interactive features to Web-based video playback, while the latter unlocks spatial video navigation opportunities. The augmented ODV principle is exposed to developers through a collection of standards-compliant Web interfaces, this way allowing for a wide range of usage scenarios and heterogeneous consumer profiles to be targeted in a broad spectrum of application domains. The paper covers the extensive set of technologies powering the concept, hereby addressing a variety of topics ranging from content authoring and editing to visualization and rendering. Two use case prototypes showcase the expressiveness and practical feasibility of the augmented ODV methodology, while performance evaluation results establish that the augmented ODV implementation is computationally sufficiently lightweight to warrant execution on low-cost PC hardware and even tablet devices.

Related Ontology Subjects/Areas/Topics (8)

Internet Technology	System Integration	Web Programming
Multimedia and User Interfaces	Web Information Systems and Technologies	Web Services and Web Engineering
Protocols and Standards	Web Interfaces and Applications

Topic(s):

Title:
ISBN:
Year:
Acronym:
Subject:

Title:
Author:
Affiliation:
Subject:

Name:
Affiliation:
Country:
Conference:
Subject:

Name:
Country:
Subject:

Affiliation

Filter listed results, by