Search

With the rapid global urbanization, several multi-dimensional complex infrastructures have emerged, introducing new challenges in the management of the vertically stratified buildings spaces. 3D indoor cadastral spaces consist a zestful research topic as their complexity and geometry alterations during time, prevents the assignment of the corresponding Rights, Restrictions and Responsibilities (RRR). In the absence of the necessary horizontal spatial data infrastructure/floor plans their determination is weak. In this paper a fit-for-purpose technical framework and a crowdsourced methodology for the implementation of 3D cadastral surveys focused on indoor cadastral spaces, is proposed and presented. As indoor data capturing tool, an open-sourced cadastral mobile application for Android devices, is selected and presented. An Indoor Positioning System based on Bluetooth technology is established while an innovative machine learning architecture is developed, in order to explore its potentials to automatically provide the position of the mobile device within an indoor environment, aiming to add more intelligence to the proposed 3D crowdsourced cadastral framework. A practical experiment for testing the examined technical solution is conducted. The produced results are assessed to be quite promising.

Energy disaggregation, or Non-Intrusive Load Monitoring (NILM), describes various processes aiming to identify the individual contribution of appliances, given the aggregate power signal. In this paper, a non-causal adaptive context-aware bidirectional deep learning model for energy disaggregation is introduced. The proposed model, CoBiLSTM, harnesses the representational power of deep recurrent Long Short-Term Memory (LSTM) neural networks, while fitting two basic properties of NILM problem which state of the art methods do not appropriately account for: non-causality and adaptivity to contextual factors (e.g., seasonality). A Bayesian-optimized framework is introduced to select the best configuration of the proposed regression model, driven by a self-training adaptive mechanism. Furthermore, the proposed model is structured in a modular way to address multi-dimensionality issues that arise when the number of appliances increases. Experimental results indicate the proposed method’s superiority compared to the current state of the art.

In January, 2020, a new virus, named SARS-COV-2, was identified and announced to the public; in March the World Health Organization (WHO) declared a worldwide pandemic. To reduce the transmissibility of the new virus, the local authorities, worldwide, introduced a series of measures to flatten the curve. Many of the measures included some form of lockdown and movement restrictions. This unique coordinated worldwide reaction, created an opportunity for researching the effects of low traffic in air quality. In this work we research the relation between the COVID-19 measures and the Air Quality Index (AQI), using four pollutant gases (CO, O3, NO2, SO2). Also, we used a variety of machine learning tools (DNN, DTR, K-NN, Lasso, LReg, MAdam, MGBR, RFR, Ridge) to estimate the accuracy of each method in the prediction of the concentration for each gas one week later. The results showed that after the strict COVID-19 restriction measures the concentration of each pollutant gas reduced rapidly and increased again after the relaxation of lockdown measures. Finally in cases like Australia, where the measures weren’t as strict as other countries, no improvement observed.

Customer Experience (CX) is monitored through market research surveys, based on metrics like the Net Promoter Score (NPS) and the customer satisfaction for certain experience attributes (e.g., call center, website, billing, service quality, tariff plan). The objective of companies is to maximize NPS through the improvement of the most important CX attributes. However, statistical analysis suggests that there is a lack of clear and accurate association between NPS and the CX attributes&rsquo, scores. In this paper, we address the aforementioned deficiency using a novel classification approach, which was developed based on logistic regression and tested with several state-of-the-art machine learning (ML) algorithms. The proposed method was applied on an extended data set from the telecommunication sector and the results were quite promising, showing a significant improvement in most statistical metrics.

Net Promoter Score (NPS) is a widely used Customer Experience Metric which when analyzed against specific attributes of experience (e.g., satisfaction associated with call center, website, etc.) leads to a classification problem with an 11×11 confusion matrix. The categorization of customers according to NPS methodology and the calculation of NPS index lead to the need to reduce the confusion matrix into a 3x3 and a 2×2 one respectively. The challenges associated with these requirements triggered the development of a generalized framework for reducing a multi-class confusion matrix into an extended 2×2 version. The method allows for the calculation of all binary classification problem metrics with potential application to any classification problem.

A combinatory approach of two well-known fields: deep learning and semi supervised learning is presented, to tackle the land cover identification problem. The proposed methodology demonstrates the impact on the performance of deep learning models when SSL approaches are used as performance functions during training. The generated soft labels, by SSL approaches, over the encoded inputs, provided by stacked encoders, allow the utilization of the entire data set during the fine-tuning step, for any deep classifier. Obtained results, at pixel level segmentation tasks over orthoimages, suggest that SSL enhanced loss functions can be beneficial in models’ performance.

An efficient, appliance-level approach for energy disaggregation, exploiting the benefits of Generative Adversarial Networks, is presented. The concept of adversarial training supports the creation of fine tuned dissagregators, which produce more detailed load estimations for a specific appliance, compared to state of the art deep learning models. The Generator and Discriminator of the model are appropriately adapted to fit the particularities of NILM problem, whereas a Seeder component is added to provide encoded compact input vectors to the Generator. The experimental evaluation against state of the art techniques indicates promising results.

Feed-in-tariff (FIT) policies are currently employed to internalize the positive externalities of renewable energy sources (RESs). However, FIT is not time varying, failing to model the dynamics of the electricity market. Thus, the concept of the aggregator has been adopted to act as a mediator between the market and RES producers. In this paper, RES aggregation is performed through virtual associations (VAs), which are dynamic clusters of prosumers created through ICT. VAs support the prosumers’ active participation in the market, the dynamic formation of the clusters to maximize prosumers’ profit and participation, and the fair competition among the VAs and among the prosumers. A VA does not have to be a separate profit-seeking entity, and thus its interests can be perfectly aligned with those of the prosumers comprising it. The fair sharing scheme used favors the most competitive VAs and prosumers, without excluding less competitive ones from the market. Different algorithms to form VAs are examined based on a min-max optimization strategy and fair sharing. Fair sharing provides: 1) incentives to the VAs to increase their competitiveness; 2) increased prosumers’ participation; and 3) dynamic interaction with the market. Experimental results obtained on realistic traces reveal the advantages of the proposed market models.

The complexity of modern urban environments and civil demands for fast, reliable and affordable decision-making requires not only a 3D Land Information System, which tends to replace traditional 2D LIS architectures, but also the need to address the time and scale parameters, that is, the 3D geometry of buildings in various time instances (4th dimension) at various levels of detail (LoDs - 5th dimension). This paper describes and proposes solutions for technical aspects that need to be addressed for the 5D modelling pipeline. Such solutions include the creation of a 3D model, the application of a selective modelling procedure between various time instances and at various LoDs, enriched with cadastral and other spatial data, and a procedural modelling approach for the representation of the inner parts of the buildings. The methodology is based on automatic change detection algorithms for spatial-temporal analysis of the changes that took place in subsequent time periods, using dense image matching and structure from motion algorithms. The selective modelling approach allows a detailed modelling only for the areas where spatial changes are detected. The procedural modelling techniques use programming languages for the textual semantic description of a building; they require the modeller to describe its part-to-whole relationships. Finally, a 5D viewer is developed, in order to tackle existing limitations that accompany the use of global systems, such as the Google Earth or the Google Maps, as visualization software. An application based on the proposed methodology in an urban area is presented and it provides satisfactory results.

Obesity comprises a common factor contributing to increased risk for a number of diseases. Research in the field has shown that a considerable percentage of obese population perceives their body figure to be significantly closer to a healthy body figure than it actually is. The current work attempts to employ image warping techniques, in combination with actual users' images, to offer a tangible view of the range of future or past stages of a body figure through the course of one or more alternative dietary/exercise plan(s) or the absence of such a plan. This functionality is offered as a module included in a web-based responsive application offering an animated view of the transformation according to input. Pure WebGL/JavaScript has been chosen as the platform to minimize footprint and maintain portability, eliminating the need for additional plugins for the visualization due to the inherent support of WebGL by most modern-day browsers.

This chapter describes the main research outcomes and achievements of 4D modelling in cultural heritage. 4D digital modelling implies the creation of precise time-varying 3D reconstructions of cultural heritage objects to capture temporal geometric variations/distortions, i.e., a spatio-temporal assessment. The key research challenge for 4D modelling, was the data collection over heterogeneous unstructured web resources. Such “in the wild” data include outliers and significant noise, since they have not been created for 3D modelling and reconstruction purposes. In addition, GPS and geo-information is limited or non-existent. However, such data allow for a massive reconstruction of the content even for monuments that have been destroyed due to natural phenomena or humans’ interventions. The key outcomes include (i) a Twitter-based 3D modelling of CH objects so as to reconstruct CH monuments and sites from unstructured image content, (ii) the development of a search engine and a (iii) recommendation system for different CH actors (curators, conservators, researchers), (iv) 3D reconstruction of the historic city of Calw in Germany, (v) the creation of a 3D virtual environment in real-time and (vi) launch of a 4D viewer enabling the easy handling of the 3D geometry plus the time. The results show the main innovation of the proposed 4D dimension, i.e., the time in precise modelling of the rich geometric content of the monuments.

In this paper, we propose a novel method to extract representative instances from choreographic sequences of 3D human motion data. The proposed key-frame extraction method implements a hierarchical scheme that exploits spatio-temporal variations of the dance movement features. The method is based on a hierarchical adaptation of the sparse modeling for representative selection algorithm (SMRS). Leveraging a joint -centric distance metric, summaries are provided at variable levels of granularity depending on the richness and complexity of the visual content at different sequence segments. The proposed method can contribute to addressing the need of organizing, indexing, archiving, retrieving and analyzing intangible (in this case, dance-related) cultural content in a tractable fashion and with lower computational and storage resource requirements. The approach is evaluated on real-world dance sequences, as well as on theatrical kinesiology datasets (available by Carnegie Mellon University). Comparisons with traditional video summarization methods show that the proposed hierarchical spatio- temporal decomposition scheme achieves promising results.

Understanding cityscapes using remote sensing data has been an active research field for more than two decades. Meanwhile, machine learning provides generalization capabilities compared to hierarchical and rule-based methods. This paper evaluates several machine learning algorithms in order to fuse shadow detection and shadow compensation methods for building detection using high resolution aerial imagery. Three complex and real-life urban study areas were used as test datasets with various: (i) kinds of buildings structures of special architecture, (ii) pixel resolutions and, (iii) types of data. Objective evaluation metrics have been used for assessing the compared algorithms such recall, precision and F1-score as well as rates of completeness, correctness and quality. For both approaches, i.e., shadow detection and building detection, the computational complexity of each machine learning algorithm was examined. The results indicate that deep learning schemes, such a Convolutional Neural Network (CNN), provides the best classification performance in terms of shadow detection and building detection.

This paper proposes a human life summarization scheme based on multimedia content published on social media. In this context the term "life" includes the events, occasions and activities users post on their walls. Towards this direction, an innovative architecture is designed that consists of two modules: the content preparation and the content summarization module. During content preparation, a Social Media web page is automatically segmented into tokens. Next multimedia content is kept and it is associated to its respective metadata (date of post, events, likes, persons, comments etc.) after filtering information through the YAGO2 knowledge base. Then a novel ranking mechanism puts multimedia content in order of importance based on a social computing methodology. Finally the summarization module produces a meaningful video clip that includes the top moments of one's life without completely disregarding the less important. To the best of the authors' knowledge, this is one of the first human life summarization schemes that are based on social media content. Experimental results illustrate the promising performance of the proposed architecture and set a basis for future research.

Determining the appropriate network topology for feedforward neural classifiers is an interesting problem that is usually addressed through evaluation against given performance metrics, such as accuracy, precision, recall, and F1-score. In this paper, we consider three more evaluation metrics inspired by the semi-supervised learning field, i.e. cluster performance, smoothness performance, and graph based metrics. Leveraging an island genetic algorithm, we investigate the applicability and efficacy of the derived multi-objective optimization strategy for defining the parameters of a feedforward neural network. The described approach is evaluated in the context of a visual classification task, but can be applied in a vast range of pattern analysis problems.

Detection of outliers in radar signals is a considerable challenge in maritime surveillance applications. High-Frequency Surface-Wave (HFSW) radars have attracted significant interest as potential tools for long-range target identification and outlier detection at over-the-horizon (OTH) distances. However, a number of disadvantages, such as their low spatial resolution and presence of clutter, have a negative impact on their accuracy. In this paper, we explore the applicability of deep learning techniques for detecting deviations from the norm in behavioral patterns of vessels (outliers) as they are tracked from an OTH radar. The proposed methodology exploits the nonlinear mapping capabilities of deep stacked autoencoders in combination with density-based clustering. A comparative experimental evaluation of the approach shows promising results in terms of the proposed methodology’s performance.

This paper introduces a decision making framework for aggregating Microgrids and/or other small energy producers and consumers (i.e. prosumers) into groups, whose purpose is to participate in liberalized electricity markets as single entities. The aggregator is able to offer aggregated Renewable Energy Source (RES) units to the wholesale market, in ways that are more efficient than individual prosumers acting alone. We first present the proposed framework and information flow among the involved market entities. We then focus on the problem of finding the set of prosumers whose aggregate prosumption profile can best fit a given target pattern requested by a market actor. We propose a linear autoregressive forecasting algorithm and a genetic clustering algorithm, which can easily adapt to the requirements set by the various use cases. Numerical results show that the aggregator can produce clusters in real time improving the average deviation from the target pattern by up to 50%.

Modelling and digitizing performing arts through motion capturing interfaces is an important aspect for the analysis, processing and documentation of intangible cultural heritage assets. However, existing modelling approaches may involve huge amounts of information which are difficult to process, store and analyze. To address these limitations, usually a skeleton describing the dancer motion is extracted. However, often the complexity still remains due to the high spatio-temporal dependencies of the detected skeleton joints. In this paper, an alternative approach is presented: choreography summarization. This means that a very small number of image frames are extracted to represent a choreography, thus significantly reducing the processing and storage complexity. In our approach the problem of choreography summarization is treated as an unsupervised clustering approach. Evaluation indices are introduced for monitoring the summarization performance. Experimental results on real-life dancing performances verifies the capability of the proposed method to capture the main patterns of a choreography with a very small number of trajectory points.

From the ancient library of Alexandria 2300 years ago, cultural collections have a common fundamental base; to gather, preserve and promote knowledge helping the intellectual and cognitive evolution of humanity. Nowadays the information revolution has given scientists, educators, researchers and individuals the ability not only to use a variety of digital libraries as a source of information but also to contribute to these libraries by uploading data that they create, leading to a massive production of digital data that we need to verify, manage, archive, preserve and reuse. Cultural heritage (CH) data is a category in digital libraries that needs our attention the most, because of their crucial role in helping us to interact with the past and learn, promote and preserve our cultural assets. Digital documentation of tangible and intangible heritage, data formats and standards, metadata and semantics, linked data, crowdsourcing and cloud, the use and reuse of data and copyright issues are the rising challenges that we try to address in this chapter, through literature research and best practice examples. At the end of this analysis, this chapter tries to predict the future of Digital Heritage Libraries, where 3D digital assets will be part of augmented, virtual and mixed reality experiences.

Resource selection and task assignment are basic operations in distributed computing environments, like the grid and the cloud, where tasks compete for resources. The decisions made by the corresponding algorithms should be judged based not only on metrics related to user satisfaction, such as the percentage of tasks served without violating their quality-of-service (QoS) requirements, but also based on resource-related performance metrics, such as the number of resources used to serve the tasks and their utilization efficiency. In our work, we focus on the case of tasks with fixed but not strict time requirements, given in the form of a requested start and finish time. We propose an algorithm for assigning tasks to resources that minimizes the violations of the tasks' time requirements while simultaneously maximizing the resources' utilization efficiency for a given number of resources. The exact time scheduling of the tasks on the resources is then decided by taking into account the time constraints. The proposed scheme exploits concepts derived from graph partitioning, and groups together tasks so as to 1) minimize the time overlapping of the tasks assigned to a given resource and 2) maximize the time overlapping among tasks assigned to different resources. The partitioning is performed using a spectral clustering methodology through normalized cuts. Experimental results show that the proposed algorithm outperforms other scheduling algorithms for different values of the granularity and the load of the task requests.

The current technological evolutions enter 3D geo-informatics into their digital age, enabling new potential applications in the field of virtual tourism, pleasure, entertainment and cultural heritage. It is argued that 3D information provides the natural way of navigation. However, personalization is a key aspect in a navigation system, since a route that incorporates user preferences is ultimately more suitable than the route with the shortest distance or travel time. Usually, user's preferences are expressed as a set of weights that regulate the degree of importance of the scene metadata on the route selection process. These weights, however, are defined by the users, setting the complexity to the user's side, which makes personalization an arduous task. In this paper, we propose an alternative approach in which metadata weights are estimated implicitly and transparently to the users, transferring the complexity to the system side. This is achieved by introducing a relevance feedback on-line learning strategy which automatically adjusts metadata weights by exploiting information fed back to the system about the relevance of user's preferences judgments given in a form of pair-wise comparisons. Practically implementing a relevance feedback algorithm presents the limitation that several pair-wise comparisons (samples) are required to converge to a set of reliable metadata weights. For this reason, we propose in this paper a weight rectification strategy that improves weight estimation by exploiting metadata interrelations defined through an ontology. In the sequel, a genetic optimization algorithm is incorporated to select the most user preferred routes based on a multi-criteria minimization approach. To increase the degree of personalization in 3D navigation, we have also introduced an efficient algorithm for estimating 3D trajectories around objects of interest by merging best selected 2D projected views that contain faces which are mostly preferred by the users. We have conducted simulations and comparisons with other approaches either in the field of on-line learning or route selection using objective metrics in terms of precision and recall values. The results indicate that our system yields on average a 13.76 % improvement of precision as regards the learning strategy and an improvement of 8.75 % regarding route selection. In addition, we conclude that the ontology driven weight rectification strategy can reduce the number of samples (pair-wise comparisons) required of 76 % to achieve the same precision. Qualitative comparisons have been also performed using a use case route scenario in the city of Athens.

We consider the fully automated behavior understanding through visual cues in industrial environments. In contrast to most existing work, which relies on domain knowledge to construct complex handcrafted features from inputs, we exploit a Convolutional Neural Network (CNN), which is a type of deep model and can act directly on the raw inputs, to automate the process of feature construction. Although such models are limited to handle still 2D inputs, in this paper we appropriately transform video input to incorporate temporal information into each frame. This way our model hierarchically constructs features from both spatial and temporal dimensions. We apply our model in real-world environment, on data taken from Nissan factory, and it achieves superior performance without relying on handcrafted features.

Cloud computing is essentially changing the way services are built, provided and consumed. As a paradigm building on a set of combined technologies, it enables service provision through the commoditization of IT assets and on-demand usage patterns. In the emerging era of the Future Internet, clouds aim at facilitating applications that move away from the monolithic approach into an Internet-scale one, thus exploiting information, individual offerings and infrastructures as composite services. In this paper we present an approach for selecting the services (that comprise the composite ones) in order to meet the end-to-end Quality of Service (QoS) requirements. The approach is enhanced with a relevance feedback mechanism that provides additional information with respect to the importance of the content and the service. The latter is performed in an automated way, allowing for user preferences to be considered during the service selection process. We also demonstrate the operation of the implemented approach and evaluate its effectiveness using a real-world scenario, based on a computer vision application.

Efficient management of natural disasters impose great research challenges to the current environmental crisis management systems in terms of both architecture and services. This is mainly due to the fact that a large amount of geospatial content is usually distributed, non-compliant to standards, and needs to be transmitted under a QoS guaranteed framework to support effective decision making either in case of an emergency or in advance planning. Incorporating real time capabilities in Web services, both in terms of dynamic configuration and service selection, is an open research agenda. The things get worst in geospatial context due to the huge amount of data transmitted from distributed sensors under heterogeneous platforms, making the need of synchronization an important issue. In this paper, we propose a flexible service oriented architecture for planning and decision support in environmental crisis management. The suggested architecture uses real time geospatial data sets and 3D presentation tools, integrated with added-value services, such as simulation models for assisting decision making in case of emergency. The proposed architectural framework goes beyond integration and presentation of static spatial data, to include real time middleware that is responsible for selecting the most appropriate method of the available geospatial content and service in order to satisfy the QoS requirements of users and/or application. A case study of a complete, real world implementation of the suggested framework dealing with forest fire crisis management system is also presented.

This paper proposes an integrated framework for analyzing human actions in video streams. Despite most current approaches that are just based on automatic spatiotemporal analysis of sequences, the proposed method introduces the implicit user-in-the-loop concept for dynamically mining semantics and annotating video streams. This work sets a new and ambitious goal: to recognize, model and properly use "average user's" selections, preferences and perception, for dynamically extracting content semantics. The proposed approach is expected to add significant value to hundreds of billions of non-annotated or inadequately annotated video streams existing in the Web, file servers, databases etc. Furthermore expert annotators can gain important knowledge relevant to user preferences, selections, styles of searching and perception.

In this paper, we provide the results of ongoing work in Magnet Beyond project, regarding social networking services. We introduce an integrated social networking framework through the definition or the appropriate notions and metrics. This allows one to run an evaluation study of three widely used clustering methods (k-means, hierarchical and spectral clustering) in the scope of social groups assessment and in regard to the cardinality of the profile used to assess users’ preferences. Such an evaluation study is performed in the context of our service requirements (i.e. on the basis of equal-sized group formation and of maximization of interests’ commonalities between users within each social group). The experimental results indicate that spectral clustering, due to the optimization it offers in terms of normalized cut minimization, is applicable within the context of Magnet Beyond socialization services. Regarding profile’s cardinality impact on the system performance, this is shown to be highly dependent on the underlying distribution that characterizes the frequency of user preferences appearance. Our work also incorporates the introduction of a heuristic algorithm that assigns new users that join the service into appropriate social groups, once the service has been initialized and the groups have been assessed using spectral clustering. The results clearly show that our approach is able to adhere to the service requirements as new users join the system, without the need of an iterative spectral clustering application that is computationally demanding.

In this paper, we address the issue of content search over peer-to peer networks. We use the concept of semantic proximity that exploits the commonalities of interests exhibited among peer users so as to decompose the network into semantic clusters. We initially define search entropy, as a metric indicating the average number of packets required to locate the requested content. Then, spectral clustering is used to organize the peer nodes into semantic clusters so that (a) the probability that a node locates content within its own cluster is maximized, while simultaneously; (b) the respective probability of finding this content outside this cluster is minimized. The proposed semantic partitioning algorithm is then extended into a hierarchical two-tier scheme, in which practical issues arising for the deployment of a peer-to-peer (p2p) application can be more easily addressed. After the system has been initialized, a dynamic algorithm places new users that join the p2p network into appropriately selected clusters and also handles peer departures without the need for matrix eigen decomposition process which is necessary for the assessment of the initial static partitioning. Our experimental results validate that (a) our static partitioning outperforms traditional and novel search techniques and (b) our dynamic algorithm is able to efficiently track the system's progression maintaining the search entropy close to the initially assessed levels.

This paper presents the main operations and technologies implemented in the framework of the EU funded COSMOS project. COSMOS introduces an advanced web repository which allows teachers and students to search, retrieve, access educational content and re-use educational material for creating earning activities through a specifically designed web interface incorporating innovative technological solutions. The repository is based on an IEEE LOM representation of the content which supports educational scenarios and learning activities as well. The architecture also supports tools for describing and managing digital content rights, which are interoperably represented using the Creative Commons Rights Expression Language (ccREL).

In this paper, we propose a new algorithm for fair scheduling, and we compare it to other scheduling schemes such as the earliest deadline first (EDF) and the first come first served (FCFS) schemes. Our algorithm uses a max-min fair sharing approach for providing fair access to users. When there is no shortage of resources, the algorithm assigns to each task enough computational power for it to finish within its deadline. When there is congestion, the main idea is to fairly reduce the CPU rates assigned to the tasks so that the share of resources that each user gets is proportional to the users weight. The weight of a user may be defined as the users contribution to the infrastructure or the price he is willing to pay for services or any other socioeconomic consideration. In our algorithms, all tasks whose requirements are lower than their fair share CPU rate are served at their demanded CPU rates. However, the CPU rates of tasks whose requirements are larger than their fair share CPU rate are reduced to fit the total available computational capacity in a fair manner. Three different versions of fair scheduling are adopted in this paper: the simple fair task order (SFTO), which schedules the tasks according to their respective fair completion times, the adjusted fair task order (AFTO), which refines the SFTO policy by ordering the tasks using the adjusted fair completion time, and the max-min fair share (MMFS) scheduling policy, which simultaneously addresses the problem of finding a fair task order and assigning a processor to each task based on a max-min fair sharing policy. Experimental results and comparisons with traditional scheduling schemes such as the EDF and the FCFS are presented using three different error criteria. Validation of the simulations using real experiments of tasks generated from 3D image- rendering processes is also provided. The three proposed scheduling schemes can be integrated into existing grid computing architectures.

In this paper, we propose an efficient non-linear task workload prediction mechanism incorporated with a fair scheduling algorithm for task allocation and resource management in Grid computing. Workload prediction is accomplished in a Grid middleware approach using a non-linear model expressed as a series of finite known functional components using concepts of functional analysis. The coefficient of functional components are obtained using a training set of appropriate samples, the pairs of which are estimated based on a runtime estimation model relied on a least squares approximation scheme. The advantages of the proposed non-linear task workload prediction scheme is that (i) it is not constrained by analysis of source code (analytical methods), which is practically impossible to be implemented in complicated real-life applications or (ii) it does not exploit the variations of the workload statistics as the statistical approaches does. The predicted task workload is then exploited by a novel scheduling algorithm, enabling a fair Quality of Service oriented resource management so that some tasks are not favored against others. The algorithm is based on estimating the adjusted fair completion times of the tasks for task order selection and on an earliest completion time strategy for the grid resource assignment. Experimental results and comparisons with traditional scheduling approaches as implemented in the framework of European Union funded research projects GRIA and GRIDLAB grid infrastructures have revealed the outperformance of the proposed method.

The inspection, assessment, maintenance and safe operation of the existing civil infrastructure consists one of the major challenges facing engineers today. Such work requires either manual approaches, which are slow and yield subjective results, or automated approaches, which depend upon complex handcrafted features. Yet, for the latter case, it is rarely known in advance which features are important for the problem at hand. In this paper, we propose a fully automated tunnel assessment approach; using the raw input from a single monocular camera we hierarchically construct complex features, exploiting the advantages of deep learning architectures. Obtained features are used to train an appropriate defect detector. In particular, we exploit a Convolutional Neural Network to construct high-level features and as a detector we choose to use a Multi-Layer Perceptron due to its global function approximation properties. Such an approach achieves very fast predictions due to the feedforward nature of Convolutional Neural Networks and Multi-Layer Perceptrons.

Outdoor large-scale cultural sites are mostly sensitive to environmental, natural and human made factors, implying an imminent need for a spatio-temporal assessment to identify regions of potential cultural interest (material degradation, structuring, conservation). On the other hand, in Cultural Heritage research quite different actors are involved (archaeologists, curators, conservators, simple users) each of diverse needs. All these statements advocate that a 5D modelling (3D geometry plus time plus levels of details) is ideally required for preservation and assessment of outdoor large scale cultural sites, which is currently implemented as a simple aggregation of 3D digital models at different time and levels of details. The main bottleneck of such an approach is its complexity, making 5D modelling impossible to be validated in real life conditions. In this paper, a cost effective and affordable framework for 5D modelling is proposed based on a spatial-temporal dependent aggregation of 3D digital models, by incorporating a predictive assessment procedure to indicate which regions (surfaces) of an object should be reconstructed at higher levels of details at next time instances and which at lower ones. In this way, dynamic change history maps are created, indicating spatial probabilities of regions needed further 3D modelling at forthcoming instances. Using these maps, predictive assessment can be made, that is, to localize surfaces within the objects where a high accuracy reconstruction process needs to be activated at the forthcoming time instances. The proposed 5D Digital Cultural Heritage Model (5D-DCHM) is implemented using open interoperable standards based on the CityGML framework, which also allows the description of additional semantic metadata information. Visualization aspects are also supported to allow easy manipulation, interaction and representation of the 5D-DCHM geometry and the respective semantic information. The open source 3DCityDB incorporating a PostgreSQL geo-database is used to manage and manipulate 3D data and their semantics.

Multimedia content modeling, i.e., identification of semantically meaningful entities, is an arduous task mainly due to the fact that (a) humans perceive the content using high-level concepts and (b) the subjectivity of human perception, which often interprets the same content in a different way at different times. For this reason, an efficient content management system has to be adapted to current user's information needs and preferences through an on-line learning strategy based on users’ interaction. One adaptive learning strategy is relevance feedback, originally developed in traditional text-based information retrieval systems. In this way, the user interacts with the system to provide information about the relevance of the content, which is then fed back to the system to update its performance. In this paper, we evaluate and investigate three main types of relevance feedback algorithms; the Euclidean, the query point movements and the correlation-based approaches. In the first case, we examine heuristic and optimal techniques which are based either on the weighted or the generalized Euclidean distance. In the second case, we survey single and multipoint query movement schemes. As far as the third type is concerned, two different ways for parametrizing the normalized cross-correlation similarity metric are proposed. The first scales only the elements of the query feature vector and called query-scaling strategy, while the second scales both the query and the selected samples (query-sample scaling strategy). All the examined algorithms are evaluated using both subjective and objective criteria. Subjective evaluation is performed by depicting the best retrieved images as response of the system to a user's query. Instead, objective evaluation is obtained using standard criteria, such as the precision–recall curve and the average normalized modified retrieval rank (ANMRR). Furthermore, a newly objective criterion, called average normalized similarity metric distance is introduced which exploits the difference among the actual and ideal similarity measure among all best retrievals. Discussions and comparisons of all the aforementioned relevance feedback algorithms are presented.

In this paper, a novel relevance feedback algorithm is proposed for improving the performance of interactive content-based retrieval systems. The algorithm recursively estimates the similarity measure, which is used for data ranking in description environments where similarity-based queries are applied, using a set of relevant/irrelevant samples feedback by the user to the system so that the adjusted response is a better approximation of the current user's information needs and preferences. In particular, using concepts of functional analysis, the similarity measure is expressed as a parametric form of known monotone increasing functional components. Then, the contribution of each functional component to the similarity measure is estimated through a recursive and efficient on-line learning algorithm so that: 1) the current user's needs and preferences, as indicated by a set of selected relevant/irrelevant samples, are satisfied as much as possible, while simultaneously 2) a minimal modification of the already estimated similarity measure is accomplished. Experimental results on a large real-life database using objective evaluation criteria, such as the precision-recall curve and the average normalized modified retrieval rank (ANMRR), indicate that the proposed scheme outperforms the compared ones. In addition, the proposed algorithm requires low computational complexity and it can be implemented in a recursive way.

Modeling multimedia content by identifying semantically meaningful entities can be arduous because it's difficult to simulate human perception. However, by creating an algorithm to respond interactively to user preference, content-retrieval systems can become more efficient and easier to use.

In this paper, an unsupervised video object (VO) segmentation and tracking algorithm is proposed based on an adaptable neural-network architecture. The proposed scheme comprises: 1) a VO tracking module and 2) an initial VO estimation module. Object tracking is handled as a classification problem and implemented through an adaptive network classifier, which provides better results compared to conventional motion-based tracking algorithms. Network adaptation is accomplished through an efficient and cost effective weight updating algorithm, providing a minimum degradation of the previous network knowledge and taking into account the current content conditions. A retraining set is constructed and used for this purpose based on initial VO estimation results. Two different scenarios are investigated. The first concerns extraction of human entities in video conferencing applications, while the second exploits depth information to identify generic VOs in stereoscopic video sequences. Human face/ body detection based on Gaussian distributions is accomplished in the first scenario, while segmentation fusion is obtained using color and depth information in the second scenario. A decision mechanism is also incorporated to detect time instances for weight updating. Experimental results and comparisons indicate the good performance of the proposed scheme even in sequences with complicated content (object bending, occlusion).

The performance of a Content-Based Image Retrieval System (CBIR) depends on (a) the system's adaptability to the user's information needs, which permits different types of indexing and simultaneously reduces the subjectivity of human perception for the interpretation of the image visual content and (b) the efficient organization of the extracted descriptors, which represent the rich visual information. Both issues are addressed in this paper. Descriptor organization is performed using a fuzzy classification scheme fragmented into multidimensional classes, instead of the previous works where fuzzy histograms were created in one dimension using, for example, the feature vector norm. Multidimensionality relates the descriptors with one another and thus allows a compact and meaningful visual representation by mapping the elements of the resulted feature vectors with a physical visual interpretation. Furthermore, fuzzy classification is applied for all visual content descriptors, in contrast to the previous approaches where only color information is exploited. Two kinds of content descriptors are extracted in our case; global-based and region-based. The first refers to the global image characteristics, while the second exploits the region-based properties. Regions are obtained by applying a multiresolution implementation of the Recursive Shortest Spanning Tree (RSST) algorithm, called M-RSST in this paper. The second issue is addressed by proposing a computationally efficient relevance feedback mechanism based on an optimal weight updating strategy. The scheme relies on the cross-correlation measure, instead of the Euclidean distance which is mainly used in most relevance feedback algorithms. Cross-correlation is a normalized measure, which expresses how similar the two feature vectors are and thus it indicates a metric of their content similarity. The proposed scheme can be recursively implemented in the case of multiple feedback iterations, instead of the previous approaches. Furthermore, it provides reliable results regardless of the number of selected sample and the feature vector size improving relevance feedback performance, as compared to other approaches.

A semi-supervised deep learning paradigm is proposed for object classification/tracking. The method addresses the main difficulties of deep learning, by allowing unsupervised data to initially configure the network and then a gradient descent optimization scheme is triggered to fine tune the data. Unsupervised learning transforms the input data into smaller and more abstract forms of representations and therefore improves the stability, convergence and performance of the model. Additionally, an adaptive approach is presented in a way to allow dynamic modification of the model to the current visual conditions. Adaptation is performed by exploiting both unsupervised and supervised samples, coming by the application of a combined motion/deep learning tracker activating only at frames a decision mechanisms ascertains retraining.

The number of digital images that are available online today has reached unprecedented levels. Recent statistics showed that by the end of 2013 there were over 250 billion photographs stored in just one of the major social media sites, with a daily average upload of 300 million photos. These photos, apart from documenting personal lives, often relate to experiences in well-known places of cultural interest, throughout several periods of time. Thus from the viewpoint of Cultural Heritage professionals, they constitute valuable and freely available digital cultural content. Advances in the fields of Photogrammetry and Computer Vision have led to significant breakthroughs such as the Structure from Motion algorithm which creates 3D models of objects using their 2D photographs. The existence of powerful and affordable computational machinery enables the reconstruction not only of single structures such as artefacts, but also of entire cities. This paper presents an overview of our methodology for producing cost-effective 4D – i.e. in space and time – models of Cultural Heritage structures such as monuments and artefacts from 2D data (pictures, video) and semantic information, freely available ‘in the wild’, i.e. in Internet repositories and social media. State-of-the-art methods from Computer Vision, Photogrammetry, 3D Reconstruction and Semantic representation are incorporated in an innovative workflow with the main goal to enable historians, architects, archaeologists, urban planners and other cultural heritage professionals to reconstruct cost-effective views of historical structures out of the billions of free images floating around the web and subsequently interact with those reconstructions.

In this paper a framework for automatic online workflow recognition in industrial environments where the issue of concurrent activities rises, is presented. The framework consists of three main parts: The first part is devoted to detecting activity in specific Regions of Interest (ROIs) of the video sequence. This is effected by separating each frame into ROIs and representing the resulting subimages through feature vectors. By observing these vectors we can determine when there is action in a particular ROI. The second part of the framework lies in examining whether the detected activity corresponds to a workflow related event. This is accomplished by HMM modeling. Finally, the third part employs a string matching based technique to confirm the validity of the observed sequence of events or correct any detection or classification errors. This last step also addresses a top down approach by informing lower system levels (such as image representation or object tracking) about the errors committed. The performance of the proposed approach is thoroughly evaluated under real-life complex visual workflow understanding scenarios, in an industrial plant. The obtained results are compared and discussed.

This paper presents an artificial neural network (ANN) approach to predicting and classifying distribution transformer specific iron losses, i.e., losses per weight unit. The ANN is trained to learn the relationship of several parameters affecting iron losses. For this reason, the ANN learning and testing sets are formed using actual industrial measurements, obtained from previous completed transformer constructions. Data comprise grain oriented steel electrical characteristics, cores constructional parameters, quality control measurements of cores production line and transformers assembly line measurements. It is shown that an average absolute error of 2.32% has been achieved in the prediction of individual core specific iron losses and an error of 2.2% in case of transformer specific losses. This is compared with average errors of 5.7% and 4.0% in prediction of specific iron losses of individual core and transformer, respectively, obtained by the current practice applying the typical loss curve to the same data.

In this paper, we give a short summary of the papers proposed in ACMARTEMIS 2013 which is held in Barcelona Spain in conjunction with ACM Multimedia. The workshop handles the areas of features analysis both at low and high level for efficient events detection, retrieval of multimedia events and objects and video synchronization issues and also events and behavior recognition from visual data. All papers were classified into three session of a single track workshop. The first session named "Video Features and Scene Analysis" includes articles that handle low level and high level visual analysis appropriate for event detection. The second session entitled "Retrieval of Multimedia Objects/Events" applies schemes for media data retrieval and video synchronization. Finally the third session "Analysis of Visual Events" describes algorithms for detecting actions, behaviors and events in complex visual scenes.

Nowadays, there are available extremely large collections of images located on distributed and heterogeneous platforms over the web. The proliferation of billions of shared photos has outpaced the current technology for browsing such collections, but at the same time it spurred the emergence of new image retrieval techniques based not only on photos' visual information, but on geo-location tags and camera exif data. Although, additional image information may be proven very useful for preliminary image retrieval, the final retrieved result is necessary to be refined by exploiting visual information.In this paper we present a process for refining image retrieval results by exploiting and fusing two unsupervised clustering techniques: DBSCAN and spectral clustering. DBSCAN algorithm is used to remove outliers from the initially retrieved image set, and spectral clustering finalizes retrieval process by clustering together visually similar images. However, DBSCAN and spectral clustering require manual tunning of their parameters, which usually requires a priori knowledge of the dataset. To overcome this problem we developed a tuning mechanism that automatically tunes the parameters of both algorithms. For the evaluation of the proposed approach we used thousands of images from Flickr downloaded using text queries for well known cultural heritage monuments.

In this paper, a tapped delay multiclass support vector machine scheme is used for supervised job classification, based on video data taken from Nissan factory. The procedure is based on multiclass SVMs enhanced with the time dimension by incorporating additional information of n-th previous frames and allowing for user feedback when necessary. Such methodology will support the visual supervision of industrial environments by providing essential information to the supervisors and supporting their job.

Google Maps and Google Earth were milestones in the evolution of Geographic Information Systems (GIS). A new era of 3D virtual geo-referenced worlds began. The need of exploration and creation personalized virtual tours is imperative. This work implements a best viewing algorithm for presenting a 3D model by considering both semantic and geometric features of the 3D model. The camera trajectory is estimated by respecting the user's preferences using a personalized entropy-based measurement. Moreover the camera speed is adjusted properly based on the complexity of the projected view of the model.

Various smart parking solutions have recently become available. They typically utilize specific technological components in their operations including RFID and various other sensors. The information gathered using such means can be processed to provide useful indicators (e.g. parking occupancy) which can be exploited by both the operators as well as the drivers. In this paper, we propose an intelligent parking reservation management architecture which is based on an optimization strategy that exploits interval scheduling principles. The aim is to improve service quality for the drivers while increasing parking lot utilization.

Social networks need to manage and control the drift of huge amounts of information by filtering and summarizing everything, in order to ensure they satisfy users' viewing pleasure. Until now social media content has already been used in a variety of applications such as for ranking of news stories, for profiling of user preferences, even for products' recommendations. However, this type of conversational, user-generated content might be used to add value to more traditional event media, such as video. In this paper we examine the capability of automatically producing meaningful summaries of generic videos. To do so we consider EdgeRank's affinity, weight and time decay parameters and implement a CLARANS-based key-frames extraction scheme. This paper forms an initial study of a social media video abstraction service and experiments indicate its promising performance.

The recent advances in ICT boost research towards the generation of personalized Geographical Information Systems (p-GIS). It is clear that selection of a route based only on geometrical criteria, i.e., the route of the shortest distance or the minimum travel time, very rarely coincides with a “satisfactory itinerary” that respects users’ preferences, that is their desires to navigate through buildings or places of his/her own particular interest. Additionally, 3D navigation gains more popularity compared with 2D approaches especially in virtual tourist and cultural heritage applications. In a p-GIS, user’s preferences can be set manually or automatically. In an automatic architecture, user preferences are expressed as a set of weights that regulate the degree of importance on the route selection process and on line learning strategies are exploited to adjust the weights. In this paper, the on-line learning strategy exploits information fed back to the system about the relevance of user’s preferences judgments given in a form of pair-wise comparisons. Then, we use a constraint fusion methodology for the dynamic modeling of user’s preference in a 3D navigation system. The method exploits an active inductive learning approach that is combined with an adaptive spectral clustering scheme in order to avoid smoothing during the weight adjustment process.