Filters
total: 6357
-
Catalog
- Publications 4287 available results
- Journals 278 available results
- Conferences 45 available results
- People 144 available results
- Inventions 2 available results
- Projects 10 available results
- Research Equipment 1 available results
- e-Learning Courses 122 available results
- Events 16 available results
- Open Research Data 1452 available results
displaying 1000 best results Help
Search results for: MULTI-TASK LEARNING, INSTRUMENT SEGMENTATION, VIDEO DEBLURRING, DENTAL MICROSCOPE, SPATIO-TEMPORAL FEATURES
-
A new multi-process collaborative architecture for time series classification
PublicationTime series classification (TSC) is the problem of categorizing time series data by using machine learning techniques. Its applications vary from cybersecurity and health care to remote sensing and human activity recognition. In this paper, we propose a novel multi-process collaborative architecture for TSC. The propositioned method amalgamates multi-head convolutional neural networks and capsule mechanism. In addition to the discovery...
-
Closer Look at the Uncertainty Estimation in Semantic Segmentation under Distributional Shift
PublicationWhile recent computer vision algorithms achieve impressive performance on many benchmarks, they lack robustness - presented with an image from a different distribution, (e.g. weather or lighting conditions not considered during training), they may produce an erroneous prediction. Therefore, it is desired that such a model will be able to reliably predict its confidence measure. In this work, uncertainty estimation for the task...
-
Visual Features for Endoscopic Bleeding Detection
PublicationAims: To define a set of high-level visual features of endoscopic bleeding and evaluate their capabilities for potential use in automatic bleeding detection. Study Design: Experimental study. Place and Duration of Study: Department of Computer Architecture, Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, between March 2014 and May 2014. Methodology: The features have...
-
Musical Instrument Tagging Using Data Augmentation and Effective Noisy Data Processing
PublicationDeveloping signal processing methods to extract information automatically has potential in several applications, for example searching for multimedia based on its audio content, making context-aware mobile applications (e.g., tuning apps), or pre-processing for an automatic mixing system. However, the last-mentioned application needs a significant amount of research to reliably recognize real musical instruments in recordings....
-
Urban scene semantic segmentation using the U-Net model
PublicationVision-based semantic segmentation of complex urban street scenes is a very important function during autonomous driving (AD), which will become an important technology in industrialized countries in the near future. Today, advanced driver assistance systems (ADAS) improve traffic safety thanks to the application of solutions that enable detecting objects, recognising road signs, segmenting the road, etc. The basis for these functionalities...
-
Lifelong Learning Idea in Architectural Education
PublicationThe recent advances in IT and technology are forcing changes in the approach to educating society. In the 20th century, life-long learning was understood as educating adults in order to improve their occupational qualifications. Life-long learning allows the needs of the present-day world to be addressed through providing the individual with education at every stage of his/her life various forms. The search for a new model...
-
Objectivization of audio-video correlation assessment experiments
PublicationThe purpose of this paper is to present a new method of conducting an audio-visual correlation analysis employing a head-motion-free gaze tracking system. First, a review of related works in the domain of sound and vision correlation is presented. Then assumptions concerning audio-visual scene creation are shortly described. The objectivization process of carrying out correlation tests employing gaze-tracking system is outlined....
-
Video of LEGO Bricks on Conveyor Belt Dataset Series
PublicationThe dataset series titled Video of LEGO bricks on conveyor belt is composed of 14 datasets containing video recordings of a moving white conveyor belt. The recordings were created using a smartphone camera in Full HD resolution. The dataset allows for the preparation of data for neural network training, and building of a LEGO sorting machine that can help builders to organise their collections.
-
Dependence of Power Characteristics on Savonius Rotor Segmentation
PublicationSavonius rotors are large and heavy because they use drag force for propulsion. This leads to a larger investment in comparison to horizontal axis wind turbine (HAWT) rotors using lift forces. A simple construction of the Savonius rotor is preferred to reduce the production effort. Therefore, it is proposed here to use single-segment rotors of high elongation. Nevertheless, this rotor type must be compared with a multi-segment...
-
Application Of Generative Adversarial Network for Data Augmentation and Multiplication to Automated Cell Segmentation of the Corneal Endothelium
PublicationConsidering the automatic segmentation of the endothelial layer, the available data of the corneal endothelium is still limited to a few datasets, typically containing an average of only about 30 images. To fill this gap, this paper introduces the use of Generative Adversarial Networks (GANs) to augment and multiply data. By using the ``Alizarine'' dataset, we train a model to generate a new synthetic dataset with over 513k images....
-
Deep Learning
PublicationDeep learning (DL) is a rising star of machine learning (ML) and artificial intelligence (AI) domains. Until 2006, many researchers had attempted to build deep neural networks (DNN), but most of them failed. In 2006, it was proven that deep neural networks are one of the most crucial inventions for the 21st century. Nowadays, DNN are being used as a key technology for many different domains: self-driven vehicles, smart cities,...
-
Voltage Harmonics Transfer through Medium Voltage Instrument Transformers
PublicationVoltage transformers are widely used in power quality monitoring systems in medium and high voltage grids. This paper presents accuracy problems related to voltage harmonics transfer through instrument transformers. A simplified lumped-parameters wideband circuit model of the voltage transformer is proposed and verified by simulation and experimental investigations. A number of voltage transformers have been tested in the frequency...
-
Bimodal deep learning model for subjectively enhanced emotion classification in films
PublicationThis research delves into the concept of color grading in film, focusing on how color influences the emotional response of the audience. The study commenced by recalling state-of-the-art works that process audio-video signals and associated emotions by machine learning. Then, assumptions of subjective tests for refining and validating an emotion model for assigning specific emotional labels to selected film excerpts were presented....
-
Analysis of the Capability of Deep Learning Algorithms for EEG-based Brain-Computer Interface Implementation
PublicationMachine learning models have received significant attention for their exceptional performance in classifying electroencephalography (EEG) data. They have proven to be highly effective in extracting intricate patterns and features from the raw signal data, thereby contributing to their success in EEG classification tasks. In this study, we explore the possibilities of utilizing contemporary machine learning algorithms in decoding...
-
English Language Learning Employing Developments in Multimedia IS
PublicationIn the realm of the development of information systems related to education, integrating multimedia technologies offers novel ways to enhance foreign language learning. This study investigates audio-video processing methods that leverage real-time speech rate adjustment and dynamic captioning to support English language acquisition. Through a mixed-methods analysis involving participants from a language school, we explore the impact...
-
Selecting Features with SVM
PublicationA common problem with feature selection is to establish how many features should be retained at least so that important information is not lost. We describe a method for choosing this number that makes use of Support Vector Machines. The method is based on controlling an angle by which the decision hyperplane is tilt due to feature selection. Experiments were performed on three text datasets generated from a Wikipedia dump. Amount...
-
Knowledge sharing and knowledge hiding in light of the mistakes acceptance component of learning culture- knowledge culture and human capital implications
PublicationPurpose: This study examines the micromechanisms of how knowledge culture fosters human capital development. Method: An empirical model was developed using the structural equation modeling method (SEM) based on a sample of 321 Polish knowledge workers employed in different industries. Findings: This study provides direct empirical evidence that tacit knowledge sharing supports human capital, whereas tacit knowledge hiding does...
-
Structural and Temporal Topic Models of Feedbacks on Service Quality – A Path to Theory Development?
PublicationThere is growing interest in applying computational methods in analysing large amount of data without sacrificing rigour in Information Systems research. In this paper, we demonstrate how the use of structural and temporal topic modelling can be employed to produce insights of both theoretical and practical importance from the analysis of textual comments on the quality of services in hospitals. As a first step, we revealed the...
-
ISSUES OF CLASSIFICATION FUNCTION CONTINUITY IN ENDOSCOPIC VIDEO CLASSIFICATION
PublicationIn the article a new way of analyzing the properties of feature vector functions (FVF) and classiers of images in a video stream is proposed. The general idea is based on focusing of the perceived continuity of the FVF and classier functions. Issues related to creating an exact mathematical model are discussed and a simplied solution is proposed. An exemplary algorithm is evaluated on three exemplary video sequences. The acquired...
-
A Method of MOS Evaluation for Video Based Services
PublicationThis paper deals with a method for QoE evaluation for the services transmitting large amount of data perceived by the end user in relatively short time periods, e.g. streaming video in mobile operator...
-
MANAGING LEARNING PROCESS WITH E-LEARNING TOOL
PublicationThis article presents one possibility to employ Moodle, the free e-Leaning platform, to organize learning understood as a process. Behavioral approach and application to massive courses are assumed. A case study is presented, where the introduction of Moodle resulted in better student performance in homework
-
Model-free and Model-based Reinforcement Learning, the Intersection of Learning and Planning
PublicationMy doctoral dissertation is intended as the compound of four publications considering: structure and randomness in planning and reinforcement learning, continuous control with ensemble deep deterministic policy gradients, toddler-inspired active representation learning, and large-scale deep reinforcement learning costs.
-
Divide and not forget: Ensemble of selectively trained experts in Continual Learning
PublicationClass-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixture-of-expert technique, where different models work together to solve the task. However, the experts are usually trained all at once using whole task data, which makes them all prone to forgetting and increasing computational burden. To address this limitation,...
-
Facial features extraction for color, frontal images
PublicationThe problem of facial characteristic features extraction is discussed. Several methods of features extraction for color en--face photographs are discussed. The methods are based mainly on the colors features related to the specific regions of the human face. The usefulness of presented methods was tested on a database of en--face photographs consisting of 100 photographs.
-
Video analytics-based algorithm for monitoring egress from buildings
PublicationA concept and a practical implementation of the algorithm for detecting of potentially dangerous situations related to crowding in passages is presented. An example of such a situation is a crush which may be caused by an obstructed pedestrian pathway. The surveillance video camera signal analysis performed in the online mode is employed in order to detect hold-ups near bottlenecks like doorways or staircases. The details of the...
-
Parallel implementation of background subtraction algorithms for real-time video processing on a supercomputer platform
PublicationResults of evaluation of the background subtraction algorithms implemented on a supercomputer platform in a parallel manner are presented in the paper. The aim of the work is to chose an algorithm, a number of threads and a task scheduling method, that together provide satisfactory accuracy and efficiency of a real-time processing of high resolution camera images, maintaining the cost of resources usage at a reasonable level. Two...
-
Video Analytics-Based Algorithm for Monitoring Egress from Buildings
PublicationA concept and practical implementation of the algorithm for detecting of potentially dangerous situations of crowding in passages is presented. An example of such situation is a crush which may be caused by obstructed pedestrian pathway. Surveillance video camera signal analysis performed on line is employed in order to detect hold-ups near bottlenecks like doorways or staircases. The details of implemented algorithm which uses...
-
Vehicle detector training with labels derived from background subtraction algorithms in video surveillance
PublicationVehicle detection in video from a miniature station- ary closed-circuit television (CCTV) camera is discussed in the paper. The camera provides one of components of the intelligent road sign developed in the project concerning the traffic control with the use of autonomous devices being developed. Modern Convolutional Neural Network (CNN) based detectors need big data input, usually demanding their manual labeling. In the presented...
-
Enterprise Gamification - Learning as a Side Effect of Competition
PublicationGmification in companies can be used for driving desired employees behaviour that are advantageous to their development and performance improvement. This paper presents tools acquired from online social networking services and game mechanisms to encourage managers to compete by providing extended statistics and user profiles features in e-learning system.
-
Selected Features of Dynamic Voting
PublicationIn multi-agent systems composed of autonomous agents with local knowledge, it is often desirable to aggregate their knowledge in order to make an educated decision. One of the methods of agreeing to a common decision is voting. A new history-based dynamic weight protocol allows for identification of the agents which contribute to the correct system decision. The main advantage of this approach, compared to static weighted system...
-
Task Scheduling – Review of Algorithms and Analysis of Potential Use in a Biological Wastewater Treatment Plant
PublicationThe idea of task scheduling is to increase the efficiency of a system by minimising wasted time, evenly loading machines, or maximising the throughput of machines. Moreover, the use of appropriate scheduling algorithms often leads to a reduction in the energy costs of the process. Task scheduling problems are found in a variety of industrial areas, and their scale changes significantly depending on the problem. This review shows...
-
Towards Cognitive and Perceptive Video Systems
PublicationIn this chapter we cover research and development issues related to smart cameras. We discuss challenges, new technologies and algorithms, applications and the evaluation of today’s technologies. We will cover problems related to software, hardware, communication, embedded and distributed systems, multi-modal sensors, privacy and security. We also discuss future trends and market expectations from the customer’s point of view.
-
Analyzing content of tasks in Business Process Management. Blending task execution and organization perspectives
PublicationAn efficient organization, management, and execution of tasks are central for the successful functioning of any organization. This topic was on the research agenda already in the early 1950s and keeps attracting the scientific community's attention today. Continuous advances and penetration of technologies in organizations are expected to increase task variety and complexity. This creates a constant demand for new methods to analyze,...
-
METHOD OF TRAINING THE ENDOSCOPIC VIDEO ANALYSIS ALGORITHMS TO MAXIMIZE BOTH ACCURACY AND STABILITY
PublicationIn the article a new training and testing method of endoscopic video analysis algorithms is presented. Classical methods take into account only eciency of recognizing objects on single video frames. Proposed method additionally considers stability of classiers output for real video input. The method is simple and can be trained on data sets created for other solutions. Therefore, it is easily applicable to existing endoscopic video...
-
A video monitoring system using ontology-driven identification of threats
PublicationIn this paper, we present a video monitoring systemthat leverages image recognition and ontological reasoningabout threats. In the solution, an image processing subsystemuses video recording of a monitored area and recognizesknown concepts in scenes. Then, a reasoning subsystem uses anontological description of security conditions and informationfrom image recognition to check if a violation of a conditionhas occurred. If a threat...
-
Limitations of Emotion Recognition from Facial Expressions in e-Learning Context
PublicationThe paper concerns technology of automatic emotion recognition applied in e-learning environment. During a study of e-learning process the authors applied facial expressions observation via multiple video cameras. Preliminary analysis of the facial expressions using automatic emotion recognition tools revealed several unexpected results, including unavailability of recognition due to face coverage and significant inconsistency...
-
Visual Features for Improving Endoscopic Bleeding Detection Using Convolutional Neural Networks
PublicationThe presented paper investigates the problem of endoscopic bleeding detection in endoscopic videos in the form of a binary image classification task. A set of definitions of high-level visual features of endoscopic bleeding is introduced, which incorporates domain knowledge from the field. The high-level features are coupled with respective feature descriptors, enabling automatic capture of the features using image processing methods....
-
MagMax: Leveraging Model Merging for Seamless Continual Learning
PublicationThis paper introduces a continual learning approach named MagMax, which utilizes model merging to enable large pre-trained models to continuously learn from new data without forgetting previously acquired knowledge. Distinct from traditional continual learning methods that aim to reduce forgetting during task training, MagMax combines sequential fine-tuning with a maximum magnitude weight selection for effective knowledge integration...
-
Speech Analytics Based on Machine Learning
PublicationIn this chapter, the process of speech data preparation for machine learning is discussed in detail. Examples of speech analytics methods applied to phonemes and allophones are shown. Further, an approach to automatic phoneme recognition involving optimized parametrization and a classifier belonging to machine learning algorithms is discussed. Feature vectors are built on the basis of descriptors coming from the music information...
-
Multi Queue Approach for Network Services Implemented for Multi Core CPUs
PublicationMultiple core processors have already became the dominant design for general purpose CPUs. Incarnations of this technology are present in solutions dedicated to such areas like computer graphics, signal processing and also computer networking. Since the key functionality of network core components is fast package servicing, multicore technology, due to multi tasking ability, seems useful to support packet processing. Dedicated...
-
Deep Learning: A Case Study for Image Recognition Using Transfer Learning
PublicationDeep learning (DL) is a rising star of machine learning (ML) and artificial intelligence (AI) domains. Until 2006, many researchers had attempted to build deep neural networks (DNN), but most of them failed. In 2006, it was proven that deep neural networks are one of the most crucial inventions for the 21st century. Nowadays, DNN are being used as a key technology for many different domains: self-driven vehicles, smart cities,...
-
Framing the Temporal Dimensions of a Brand
PublicationDrawing on existing research dealing with time in brand and brand management, this paper aims at providing a comprehensive and coherent framework of some time-related concepts, with a special emphasis on what happens when a brand reaches the senescence stage. In addition, it strives to consider what happens when a brand becomes long-lived enough, looking at the brand’s customer base. While undoubtedly time affects customers’ age...
-
Miscanthus × giganteus light microscope observation - short and double rotation
Open Research DataThis dataset includes light microscope images of cross root, leaf, and rhizome sections from Miscanthus × giganteus. The experiment was performed under work package 2: Energy plantation - field trial, research task 1: Energy biomass cultivation in short and double rotation. The plants were planted in marginal soils that had been amended with biochar...
-
Oral Health-Related Knowledge, Attitudes and Behaviours of Arab Dental Students: Multi-National Cross-Sectional Study and Literature Analysis 2000–2020
Publication -
Cartographic Representation of Route Reconstruction Results in Video Surveillance System
PublicationThe video streams available in a surveillance system distributed on the wide area may be accompanied by metadata are obtained as a result of video processing. Many algorithms applied to surveillance systems, e.g. event detection or object tracking, are strictly connected with localization of the object and reconstruction of its route. Drawing related information on a plan of a building or on a map of the city can facilitate the...
-
Thermographic images during testing measuring instrument boards
Open Research DataThe dataset presents thermografic images acquired during testing the prototype of measuring instrument for resistiometric corrosion monitoring. The testing was performed in the room temperature. VIGOcam v50 thermal imaging camera (VIGO System S.A., Ozarow Mazowiecki, Poland) was used for taking the pictures. These pictures were used as preliminary tests...
-
Detection of moving objects in images combined from video and thermal cameras
PublicationAn algorithm for detection of moving objects in video streams from the monitoring cameras is presented. A system composed of a standard video camera and a thermal camera, mounted in close proximity to each other, is used for object detection. First, a background subtraction is performed in both video streams separately, using the popular Gaussian Mixture Models method. For the next processing stage, the authors propose an algorithm...
-
Deep Features Class Activation Map for Thermal Face Detection and Tracking
PublicationRecently, capabilities of many computer vision tasks have significantly improved due to advances in Convolutional Neural Networks. In our research, we demonstrate that it can be also used for face detection from low resolution thermal images, acquired with a portable camera. The physical size of the camera used in our research allows for embedding it in a wearable device or indoor remote monitoring solution for elderly and disabled...
-
Parallelization of video stream algorithms in kaskada platform
PublicationThe purpose of this work is to present different techniques of video stream algorithms parallelization provided by the Kaskada platform - a novel system working in a supercomputer environment designated for multimedia streams processing. Considered parallelization methods include frame-level concurrency, multithreading and pipeline processing. Execution performance was measured on four time-consuming image recognition algorithms,...
-
Eigenfaces, Fisherfaces, Laplacianfaces, Marginfaces – How to Face the Face Verification Task
PublicationThis paper describes the exhaustive tests of four known methods of linear transformations (Eigenfaces, Fisherfaces, Laplacianfaces and Marginfaces) in the context of face verification task. Additionally, we introduce a new variant of the transformation (Laplacianface + LDA), and the specific interval-based decision rule. Both of them improve the performance of face verification, in general, however, our experiments show that the...