Papers by Zenonas Theodosiou
Visual-Semantic Context Learning for Image Classification
2016 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), 2016
With the emerging and intense use of Online Social Networks (OSNs) amongst young children and tee... more With the emerging and intense use of Online Social Networks (OSNs) amongst young children and teenagers (youngsters), safe networking and socializing on the Web has faced extensive scrutiny. Content and interactions which are considered safe for adult OSN users might embed potentially threatening and malicious information when it comes to underage users. This work is motivated by the strong need to safeguard youngsters OSNs experience such that they can be empowered and aware. The topology of a graph is studied towards detecting the so called "social bridges", i.e. the major supporters of malicious users, who have links and ties to both honest and malicious user communities. A graph-topology based classification scheme is proposed to detect such bridge linkages which are suspicious for threatening youngsters networking. The proposed scheme is validated by a Twitter network, at which potentially dangerous users are identified based on their Twitter connections. The achieved performance is higher compared to previous efforts, despite the increased complexity due to the variety of groups identified as malicious.
Fluorescent in-situ hybridization (FISH) and immunohistochemistry (IHC) constitute a pair of comp... more Fluorescent in-situ hybridization (FISH) and immunohistochemistry (IHC) constitute a pair of complimentary techniques for detecting gene amplification and overexpression, respectively. The advantages of IHC include relatively cheap materials and high sample durability, while FISH is the more accurate and reproducible method. Evaluation of FISH and IHC images is still largely performed manually, with automated or semiautomated techniques increasing in popularity. Here, we provide a comprehensive review of a number of (semi-) automated FISH and IHC image processing systems, focusing on the algorithmic aspects of each technique. Our review verifies the increasingly important role of such methods in FISH and IHC; however, manual intervention is still necessary in order to resolve particularly challenging or ambiguous cases. In addition, large-scale validation is required in order for these systems to enter standard clinical practice.
Proceedings of the 31st ACM International Conference on Information & Knowledge Management
It is increasingly easy for interested parties to play a role in the development of predictive al... more It is increasingly easy for interested parties to play a role in the development of predictive algorithms, with a range of available tools and platforms for building datasets, as well as for training and evaluating machine learning (ML) models. For this reason, it is essential to create awareness among practitioners on the ethical challenges, such as the presence of social bias in training data. We present RECANT (Raising Awareness of Social Bias in Crowdsourced Training Data), a tool that allows users to explore the behaviors of four biometric models-predicting the gender and race, as well as the perceived attractiveness and trustworthiness, of the person depicted in an input image. These models have been trained on a crowdsourced dataset of passport-style people images, where crowd annotators described attributes of the images, and reported their own demographic characteristics. With RECANT, users can explore the correct and wrong predictions made by each model, when using different subsets of the data in training, based on annotator attributes. We present its features, along with sample exercises, as a hands-on tool for raising awareness of potential pitfalls in data practices surrounding ML.
Digital Heritage. Progress in Cultural Heritage: Documentation, Preservation, and Protection, 2021
The ability to identify the artworks that a museum visitor is looking at, using first-person imag... more The ability to identify the artworks that a museum visitor is looking at, using first-person images seamlessly captured by wearable cameras can be used as a means for invoking applications that provide information about the exhibits, and provide information about visitors' activities. As part of our efforts to optimize the artwork recognition accuracy of an artwork identification system under development, an investigation aiming to determine the effect of different conditions on the artwork recognition accuracy in a gallery/exhibition environment is presented. Through the controlled introduction of different distractors in a virtual museum environment, it is feasible to assess the effect on the recognition performance of different conditions. The results of the experiment are important for improving the robustness of artwork recognition systems, and at the same time the conclusions of this work can provide specific guidelines to curators, museum professionals and visitors, that will enable the efficient identification of artworks, using images captured with wearable cameras in a museum environment.
Detection and Recognition of Barriers in Egocentric Images for Safe Urban Sidewalks
Communications in Computer and Information Science, 2022
Image Retrieval Using Keywords: The Machine Learning Perspective
Digital Imaging and Computer Vision, 2014
This chapter focuses on image retrieval using keywords under the perspec- tive of machine learnin... more This chapter focuses on image retrieval using keywords under the perspec- tive of machine learning. It covers different aspects of the current research in this area, including low-level feature extraction, creation of training sets and development of machine learning methodologies. It also presents the evalua- tion framework of automatic image annotation and discusses various methods and metrics utilized within it. Furthermore, it proposes the idea of addressing automatic image annotation by creating visual models, one for each available keyword, and presents an example of the proposed idea by comparing differ- ent features and machine learning algorithms in creating visual models for keywords referring to the athletics domain.
Algorithms
The provision of information encourages people to visit cultural sites more often. Exploiting the... more The provision of information encourages people to visit cultural sites more often. Exploiting the great potential of using smartphone cameras and egocentric vision, we describe the development of a robust artwork recognition algorithm to assist users when visiting an art space. The algorithm recognizes artworks under any physical museum conditions, as well as camera point of views, making it suitable for different use scenarios towards an enhanced visiting experience. The algorithm was developed following a multiphase approach, including requirements gathering, experimentation in a virtual environment, development of the algorithm in real environment conditions, implementation of a demonstration smartphone app for artwork recognition and provision of assistive information, and its evaluation. During the algorithm development process, a convolutional neural network (CNN) model was trained for automatic artwork recognition using data collected in an art gallery, followed by extensive ...
Biometrics is an automated authentication mechanism that allows the identification or verificatio... more Biometrics is an automated authentication mechanism that allows the identification or verification of individual based on unique physiological and behavioural characteristics. The combination of two or more biometric technologies in one application, better known as a multimodal biometric system, can provide enhanced security. Apart from the sound choice of fusion methodologies for the combination of single modality biometrics, the success of such multimodal biometric systems significantly relies on the availability of biometric databases, through which the validation of these systems is made possible. This paper presents a new multimodal database, acquired in the framework of the POLYBIO project funded by the Cyprus Research Promotion Foundation (CRPF). The database consists of fingerprint images captured via an optical sensor, frontal and side views of still and video face images as well as the outside surface of the human palm from two web-camera sensors, and a series of voice utterances recorded with the use of a distant array microphone. The POLYBIO database includes real multimodal and multi-biometric data from 45 individuals acquired in just a single session. In this contribution, the novel platform for data acquisition and combination-through an integrated device-of the four aforementioned single biometric modalities is described and the protocols used for this purpose as well as the contents of the database and its statistics are presented.
Encouraging people to walk rather than using other means of transportation is an important factor... more Encouraging people to walk rather than using other means of transportation is an important factor towards personal health and environmental sustainability. However, given the large number of pedestrian accidents recorded every year, the need for safe urban environments is increasing. Taking advantage of the potential of citizen-science for crowdsourcing data and creating awareness, we developed a smartphone application for enhancing the safety of pedestrians while walking in cities. Using the application, citizens will monitor the urban sidewalks and update a crowdsourcing platform with the detected barriers and damages that hinder safe walking, along with their location on a city map. To help users assign the correct type of obstacle, and authorities to assess the urgency, a Convolutional Neural Network (CNN) model for barrier and damage recognition is embedded in the application. The results of a user evaluation, based on a group of volunteers who used the application in real conditions, demonstrate the potential of using the application in conjunction with a smart city framework.
HER2-positive breast cancer is characterized by aggressive growth and poor prognosis. Women with ... more HER2-positive breast cancer is characterized by aggressive growth and poor prognosis. Women with metastatic breast cancer with over-expression of HER2 protein or excessive presence of HER2 gene copies are potential candidates for Herceptin (Trastuzumab) targeted treatment that binds to HER2 receptors on tumor cells and inhibits tumor cell growth. Fluorescence in situ hybridization (FISH) is one of the most widely used methods to determine HER2 status. Typically, evaluation of FISH images involves manual counting of FISH signals in multiple images, a time consuming and error prone procedure. Recently, we developed novel software for the automated evaluation of FISH images and, in this study, we present the first testing of this software on images from two separate research clinics. To our knowledge, this is the first concurrent evaluation of any FISH image analysis software in two different clinics. The evaluation shows that the developed FISH image analysis software can accelerate e...
Companion Proceedings of the The Web Conference 2018, 2018
As smart cities infrastructures mature, data becomes a valuable asset which can radically improve... more As smart cities infrastructures mature, data becomes a valuable asset which can radically improve city services and tools. Registration, acquisition and utilization of data, which will be transformed into smart services, are becoming more necessary than ever. Online social networks with their enormous momentum are one of the main sources of urban data offering heterogeneous real-time data at a minimal cost. However, various types of attacks often appear on them, which risk users' privacy and affect their online trust. The purpose of this article is to investigate how risks on online social networks affect smart cities and study the differences between privacy and security threats with regard to smart people and smart living dimensions.
HER2/neu gene amplification is being evaluated by fluorescent in situ hybridization (FISH). In or... more HER2/neu gene amplification is being evaluated by fluorescent in situ hybridization (FISH). In order to avoid interobserver variations in the assessment of HER2/neu status, an integrated FISH image analysis system is developed to automate the classification of FISH images from breast carcinomas. Using a two-stage algorithm, for nuclei and dot detection, and combining results from multiple images taken from a slice for overall case classification, FISH signals ratio per cell nucleus were measured and cases were classified as positive or negative. The system consists of functions for red spot detection, green spot detection, nuclei segmentation and FISH signal ratio. Therefore, it provides the capability to manually correct the resulted images after the analysis.
Crowdsourcing annotation is a recent development since a complete and elaborate annotation of the... more Crowdsourcing annotation is a recent development since a complete and elaborate annotation of the content of an image is an extremely labour-intensive and time consuming task. In this paper we examine the possibility to build accurate visual models for keywords created through crowdsourcing. Specifically, 8 different keywords related to athletics domain have been modelled using MPEG-7 and Histogram of Oriented Gradients (HOG) low level features and the Sequential Minimal Optimization (SMO) classifier. The experimental results have been examined using accuracy metrics and are very promising showing the ability of the visual models to classify the images into the 8 classes with the highest average accuracy rate of 73.13% in the purpose of the HOG features.
Visual Lifelogs Retrieval: State of the Art and Future Challenges
2019 14th International Workshop on Semantic and Social Media Adaptation and Personalization (SMAP), 2019
The use of wearable cameras covers several areas of application nowadays, where the need for deve... more The use of wearable cameras covers several areas of application nowadays, where the need for developing smart applications providing the sustainability and well-being of citizens it is more necessary than ever before. The tremendous amount of lifelogging data to extract valuable knowledge about the every day life of the wearers requires state of the art retrieval techniques to efficiently store, access, search and retrieve useful information. Several works have been proposed combining computer vision and machine learning techniques to analyze the content of the data captured from visual wearable devices on a daily basis. This paper presents an overview of the progress in visual lifelogging retrieval and indicates the current advances and future challenges, highlighting the prospects of incorporating visual lifelogging retrieval in social computing applications.
Recent advances in digital video technology have resulted in an explosion of digital video data w... more Recent advances in digital video technology have resulted in an explosion of digital video data which are available through the Web or in private repositories. Efficient searching in these repositories created the need of semantic labeling of video data at various levels of granularity, i.e., movie, scene, shot, keyframe, video object, etc. Through multilevel labeling video content is appropriately indexed, allowing access from various modalities and for a variety of applications. However, despite the huge efforts for automatic video annotation human intervention is the only way for reliable semantic video annotation. Manual video annotation is an extremely laborious process and efficient tools developed for this purpose can make, in many cases, the true difference. In this paper we present a video annotation tool, which uses structured knowledge, in the form of XML dictionaries, for semantic labeling, and an efficient shot detection algorithm for automatic video segmentation. Furth...
With the emerging and intense use of Online Social Networks (OSNs) amongst young children and tee... more With the emerging and intense use of Online Social Networks (OSNs) amongst young children and teenagers (youngsters), safe networking and socializing on the Web has faced extensive scrutiny. Content and interactions which are considered safe for adult OSN users might embed potentially threatening and malicious information when it comes to underage users. This work is motivated by the strong need to safeguard youngsters OSNs experience such that they can be empowered and aware. The topology of a graph is studied towards detecting the so called “social bridges”, i.e. the major supporters of malicious users, who have links and ties to both honest and malicious user communities. A graph-topology based classification scheme is proposed to detect such bridge linkages which are suspicious for threatening youngsters networking. The proposed scheme is validated by a Twitter network, at which potentially dangerous users are identified based on their Twitter connections. The achieved performan...
Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2020
Egocentric vision, which relates to the continuous interpretation of images captured by wearable ... more Egocentric vision, which relates to the continuous interpretation of images captured by wearable cameras, is increasingly being utilized in several applications to enhance the quality of citizens' life, especially for those with visual or motion impairments. The development of sophisticated egocentric computer vision techniques requires automatic analysis of large databases of first-person point of view visual data collected through wearable devices. In this paper, we present our initial findings regarding the use of wearable cameras for enhancing the pedestrians' safety while walking in city sidewalks. For this purpose, we create a first-person database that entails annotations on common barriers that may put pedestrians in danger. Furthermore, we derive a framework for collecting visual lifelogging data and define 24 different categories of sidewalk barriers. Our dataset consists of 1796 annotated images covering 1969 instances of barriers. The analysis of the dataset by means of object classification algorithms, depict encouraging results for further study.
Smart Cities at Risk!
Companion of the The Web Conference 2018 on The Web Conference 2018 - WWW '18, 2018
We propose a novel supervised learning rule allowing the training of a precise input-output behav... more We propose a novel supervised learning rule allowing the training of a precise input-output behavior to a spiking neuron. A single neuron can be trained to associate (map) different output spike trains to different multiple input spike trains. Spike trains are transformed into continuous functions through appropriate kernels and then Delta rule is applied. The main advantage of the method is its algorithmic simplicity promoting its straightforward application to building spiking neural networks (SNN) for engineering problems. We experimentally demonstrate on a synthetic benchmark problem the suitability of the method for spatio-temporal classification. The obtained results show promising efficiency and precision of the proposed method.
Uploads
Papers by Zenonas Theodosiou