Anastasios Doulamis

... airplane. Nevertheless, although VOs and their functionalities are analytically described by the MPEG-4 standard, techniques to generate them are left to content developers. ... content. For this reason a shot cut detection algorithm is applied. ...

In this chapter, a non-linear relevance feedback mechanism is proposed for increasing the performance and the reliability of information (medical content) retrieval systems. In greater detail, the user who searches for information is considered to be part of the retrieval ...

ABSTRACT 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.

ABSTRACT This paper presents a vision-based system for maritime surveillance using moving PTZ cameras. This system is intended to be used as an early warning system by local authorities. It fuses a visual attention method that exploits low-level image features appropriately selected for maritime environment, with an online adaptable neural network tracker, without making any assumptions about environmental or visual conditions. Systems performance was evaluated with videos from cameras placed at Limassol port and Venetian port of Chania and concerns robustness compared to dynamically changing visual conditions and different kinds of vessels, all in real time.

ABSTRACT This article presents a robust, real-time background subtraction algorithm able to operate properly in complex dynamically changing visual conditions and indoor/outdoor environments, based on a single, cheap monocular camera, like a webcam. This algorithm uses an image grid and models each pixel of the grid as a mixture of adaptive Student-t distributions. This approach makes this algorithm robust and efficient, in terms of computational cost and memory requirements, and thus suitable for large scale implementations. The proposed algorithm is applied in the problem of humans' fall detection that presents high complexity of visual content. Finally, the performances of this scheme and the scheme proposed in [1] by the same authors, are compared.

ABSTRACT Production lines need fast technologies for critical online decisions. Orchestration of different signals and processes from a number of different sensors around the industry, into related workflows are the key signals on workflows tasks recognition. The advantages of sensor networks’ technologies outputs, that their outputs are controlled by multi agents systems, in order to orchestrate and execute related production workflows, is the main point of this research work. In specific, the aim of this manuscript is to present a system that manipulates and controls these workflows executions. System architecture and sub modules are presented in more detail through the manuscript, plus a production scenario implementation.

ABSTRACT Market simulators assist managers in product design and pricing decisions, through advanced heuristic algorithms for consumer buying behavior modeling. We propose a model for use in such systems, which combines the effective incorporation of the two critical properties a simulator should theoretically reflects, with exceptionally performance in actual choice shares estimation. Differential Impact and Substitution are exhibited through the representation of customer heterogeneity and product similarity in the choice rule, while high predictive accuracy is displayed with the implementation of the Stochastic Logarithmic Search algorithm.

ABSTRACT This paper presents a system suitable to perform precise and fast 3D measurements from synchronized stereo-video sequences and provide target’s georeference in a known reference system. To this direction we combine a robust tracker with photogrammetric techniques into a fast and reliable system. For tracking objects and people, we adopt and modify a stable human tracker able to cope efficiently with the trade-off between model stability and adaptability. For achieving accurate and precise 3D measurements, camera calibration was implemented in order to recover the intrinsic parameters of the cameras of the configuration. Finally, for precise and reliable calculation of the 3D trajectory of the moving person, we apply bundle adjustment for all frames. Bundle adjustment is a very accurate algorithm and has the advantages of being tolerant of missing data while providing a true Maximum Likelihood estimate. The results have been tested and evaluated in real life conditions for proving the robustness and the accuracy of the system.

ABSTRACT Designing optimal product lines is essential for any firm to stay competitive. Whereas a large number of optimization algorithms have been applied for solving the problem, most of them adopt the first choice rule to simulate the consumer's choice process. Researchers avoid using probabilistic choice models, such as the logit, since they tend to produce duplicate products in the line, due to the IIA problem. Furthermore preference heterogeneity among consumers is a factor usually neglected, although its representation form has a substantial impact in the design of the product line. We propose a probabilistic choice model that can be used with algorithms that solve the optimal product line design problem, using the share of choices criterion. Our model deals with the IIA problem, by incorporating the similarity among products through the use of a corrective method. In addition, preference heterogeneity among consumers is effectively represented, while the model's predictive accuracy is optimized through the use of Stochastic Logarithmic Search and Genetic Algorithms.

ABSTRACT In 2007, our planet became predominantly urban as for the first time, more than half of the world's population was living in cities. This tendency for urbanization is characterized by urban sprawl phenomena and excessive environmental pollution degrading citizens' living standards. Although intelligent and safe urban planning and management is a challenging task, latest ICT advances can offer promising solutions to the aforementioned problem by providing efficient and effective ways for handling large amounts of unstandardized and distributed information. In this paper we present an integrated, fully automatic ICT solution for acquiring, processing and representing heterogeneous, unstandardized and distributed information about underground gas pipeline networks, in order to provide a safe way for urban growth.

We present a large-scale platform for distributing Virtual Reality advertisements over the World Wide Web. The platform aims at receiving and transmitting large amounts of data over mobile and desktop devices in Smart City contexts, is based on a modular and distributed architecture to allow for scalability, and incorporates content-based search capabilities for VR scenes to allow for content management. Data is stored on a cloud repository, to allow for a large amount of VR material to be kept and distributed, and follows a service-based approach of independent subsystems for the management, conversion and streaming of information. In order to function over a wide range of used end-devices, from mobile phones to high-end desktop PCs, the system is based on HTML5 technologies, and implements a remote rendering server to alleviate the computational burden on the end device. Furthermore, an extension of the MPEG-7 standard is used for the description and retrieval of 3D scenes from the cloud, and we have further ensured compliance of our system with a number of other structure and communication standards, to ensure extensibility and reusability of the sub-modules. The platform is a research work in progress: we present the subsystems already implemented, plan our next steps and describe our contributions to research.

ABSTRACT This paper presents a vision-based system for maritime surveillance, using moving PTZ cameras. The proposed methodology fuses a visual attention method that exploits low-level image features appropriately selected for maritime environment, with appropriate tracker, without making any assumptions about environmental or visual conditions. The offline initialization is based on large graph semi-supervised technique. System’s performance was evaluated with videos from cameras placed at Limassol port and Venetian port of Chania. Results suggest high detection ability, despite dynamically changing visual conditions and different kinds of vessels, all in real time.