N. Sarris | UMIST, Manchester, England, UK (original) (raw)
Papers by N. Sarris
This paper presents a new Web tool for converting text to Sign Language notation and correspondin... more This paper presents a new Web tool for converting text to Sign Language notation and corresponding VRML animation sequences for H-anim compliant avatars. The tool can be used both as a sign language dictionary as well as a text to sign language translator. It is based on a Sign Language dictionary written in the well-known Sign-Writing system. More specifically, the entries of the dictionary are in SWML (SignWriting Markup Language), an XML-based format, which has recently been developed for the storage, indexing and processing of SignWriting notation. For sign synthesis, each sign box (basic sign) is first converted to a sequence of Body Animation Parameters (BAPs) of the MPEG-4 standard corresponding to the represented gesture. These sequences, which can also be coded and/or reproduced by MPEG-4 BAP players, are then used to animate H-anim compliant VRML avatars, reproducing the exact gestures represented in sign language notation. Envisaged applications include the development of...
Abstract. Digital archives comprise a valuable asset for effective information retrieval. In many... more Abstract. Digital archives comprise a valuable asset for effective information retrieval. In many cases, however, the special vocabulary of the archive restricts its access only to experts in the domain of the material it contains and, as a result, researchers of other disciplines or the general public cannot take full advantage of the wealth of information it offers. To this end, the Papyrus research project has worked towards a solution which makes cross-discipline search possible in digital libraries. The developed prototype showcases this approach demonstrating how we can discover history in news archives. In this demo we focus on demonstrating two of the end user tools available in the prototype, the cross-discipline search and the Papyrus browser.
This paper describes the 'LipTelephone' system, which is being developed under a Greek AM... more This paper describes the 'LipTelephone' system, which is being developed under a Greek AMEA research project, with the purpose to serve as a videophone that can also be used by lip readers. This system aims to combine model-based with traditional coding techniques in order to exploit the information redundancy in a scene of known content, while achieving high fidelity representation in the specific area of interest, which is the speaker's mouth. The recent MPEG-4 coding standard provides the framework, which is employed, for these purposes.
Proceedings of the Conference for Truth and Trust Online 2019
This paper presents an overview of the WeVerify H2020 EU project, which is developing intelligent... more This paper presents an overview of the WeVerify H2020 EU project, which is developing intelligent human-in-the-loop content verification and disinformation analysis methods. Social media and web content are analysed and contextualised within the broader online ecosystem, in order to expose fabricated content, through cross-modal content verification, social network analysis, microtargeted debunking and a blockchain-based public database of known fakes. In particular, we introduce the following already developed tools: cross-modal content verification; blockchain-based database of known fakes; an open source content verification browser plugin; a collaborative verification workbench.
Proceedings of the 23rd International Conference on World Wide Web - WWW '14 Companion, 2014
ABSTRACT There are both positive and negative aspects in the use of social media in news and info... more ABSTRACT There are both positive and negative aspects in the use of social media in news and information dissemination. To deal with the negative aspects, such as the spread of rumours and fake news, the flow of information should implicitly be filtered and marked to specific criteria such as credibility, trustworthiness, reputation, popularity, influence, and authenticity. This paper proposes an approach that can enhance trustworthiness and content validity in the presence of information overload. We introduce Alethiometer, a framework for assessing truthfulness in social media that can be used by professional and general news users alike. We present different measures that delve into the detailed analysis of the content, the contributors of the content and the underlying context. We further propose an approach for deriving a single metric that considers, in a unified manner, the quality of a contributor and of the content provided by that contributor. Finally, we present some preliminary statistical results from the examination of a set of 10 million twitter users, that provide useful insights on the characteristics of social media data.
Keywords: 3D model based coding, rigid head tracking, 3D motion estimation, feature tracking The ... more Keywords: 3D model based coding, rigid head tracking, 3D motion estimation, feature tracking The presented work proposes substantial improvements to established methods for the twodimensional tracking of an image and the rigid adaptation of a three-dimensional face model to a 2D projection. These are combined to produce a robust system for the rigid 3D tracking of a human head which is promising even for extreme rotation conditions where most methods fail. Our technique, while using a 2D feature-based image tracking algorithm introduced by Kanade, Lucas and Tomasi, is based on the assessment of the tracked features considering their suitability for tracking, the success in tracking them in the current frame and their reliability judged from their 3D position in a model adapted to the previous frame. Having determined the optimum set of tracked features on the current image frame these are used to readapt the 3D human face model which has been initialised to fit to the face manually ...
There are both positive and negative aspects in the use of social media in news and information d... more There are both positive and negative aspects in the use of social media in news and information dissemination. To deal with the negative aspects, such as the spread of rumours and fake news, the flow of information should implicitly be filtered and marked to specific criteria such as credibility, trustworthiness, reputation, popularity, influence, and authenticity. This paper proposes an approach that can enhance trustworthiness and content validity in the presence of information overload. We introduce Alethiometer, a framework for assessing truthfulness in social media that can be used by professional and general news users alike. We present different measures that delve into the detailed analysis of the content, the contributors of the content and the underlying context. We further propose an approach for deriving a single metric that considers, in a unified manner, the quality of a contributor and of the content provided by that contributor. Finally, we present some preliminary s...
The tourism & leisure-related information contents provided by the services identified at the ASK... more The tourism & leisure-related information contents provided by the services identified at the ASK-IT pilot sites have about the same quality as the examined site-independent best-practice examples. The observations reported in the previous section are hence also valid for the siterelated services: The services focus on specific user groups while neglecting others. The service contents show generally also the same bias toward user groups with physical impairments.
� Abstract— This is a brief description of MESH - Multimedia semantic syndication for enhanced ne... more � Abstract— This is a brief description of MESH - Multimedia semantic syndication for enhanced news services (http://www.mesh-ip.eu), an EC co-funded project that started in March 2006 to create a framework for intelligent creation and delivery of semantically-enhanced multimedia news.
International Workshop on Intelligent Communication Technologies and Applications with Emphasis on Mobile Communications, May 5, 1999
The presented work proposes substantial improvements to established methods for the twodimensiona... more The presented work proposes substantial improvements to established methods for the twodimensional tracking of an image and the rigid adaptation of a three-dimensional face model to a 2D projection. These are combined to produce a robust system for the rigid 3D tracking of a human head which is promising even for extreme rotation conditions where most methods fail. Our technique, while using a 2D feature-based image tracking algorithm introduced by Kanade, Lucas and Tomasi, is based on the assessment of the tracked ...
Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101), 2000
This paper addresses the problem of adapting a generic 3D face model to a human face of which the... more This paper addresses the problem of adapting a generic 3D face model to a human face of which the frontal and profile views are given. Assuming that a set of feature points have been detected on both views the adaptation procedure initializes with a rigid transformation of the model aiming to minimize the distances of the 3D model feature nodes from the calculated 3D coordinates of the 2D feature points. Then, a non-rigid transformation ensures that the feature nodes are displaced optimally close to their exact calculated positions, dragging their neighbors in a way that does not deform the facial model in an unnatural way.
Technology, Management and Applications, 2002
Proceedings of the 5th Seminar on Neural Network Applications in Electrical Engineering. NEUREL 2000 (IEEE Cat. No.00EX287), 2000
Proceedings of International Conference on Image Processing, 1997
A coding method for videophone communications over wireless, varying bit-rate, channels, is propo... more A coding method for videophone communications over wireless, varying bit-rate, channels, is proposed. This method is based on defining one or more areas of interest and breaking them into a number of arbitrarily shaped regions which are homogeneous in their motion. The interest areas in the current implementation are human faces. They are detected with the help of a 2D
Lecture Notes in Computer Science, 2011
Digital archives comprise a valuable asset for effective information retrieval. In many cases, ho... more Digital archives comprise a valuable asset for effective information retrieval. In many cases, however, the special vocabulary of the archive restricts its access only to experts in the domain of the material it contains and, as a result, researchers of other disciplines or the general public cannot take full advantage of the wealth of information it offers. To this end, the Papyrus research project has worked towards a solution which makes cross-discipline search possible in digital libraries. The developed prototype showcases this approach demonstrating how we can discover history in news archives. In this demo we focus on demonstrating two of the end user tools available in the prototype, the cross-discipline search and the Papyrus browser.
This paper presents a new Web tool for converting text to Sign Language notation and correspondin... more This paper presents a new Web tool for converting text to Sign Language notation and corresponding VRML animation sequences for H-anim compliant avatars. The tool can be used both as a sign language dictionary as well as a text to sign language translator. It is based on a Sign Language dictionary written in the well-known Sign-Writing system. More specifically, the entries of the dictionary are in SWML (SignWriting Markup Language), an XML-based format, which has recently been developed for the storage, indexing and processing of SignWriting notation. For sign synthesis, each sign box (basic sign) is first converted to a sequence of Body Animation Parameters (BAPs) of the MPEG-4 standard corresponding to the represented gesture. These sequences, which can also be coded and/or reproduced by MPEG-4 BAP players, are then used to animate H-anim compliant VRML avatars, reproducing the exact gestures represented in sign language notation. Envisaged applications include the development of...
Abstract. Digital archives comprise a valuable asset for effective information retrieval. In many... more Abstract. Digital archives comprise a valuable asset for effective information retrieval. In many cases, however, the special vocabulary of the archive restricts its access only to experts in the domain of the material it contains and, as a result, researchers of other disciplines or the general public cannot take full advantage of the wealth of information it offers. To this end, the Papyrus research project has worked towards a solution which makes cross-discipline search possible in digital libraries. The developed prototype showcases this approach demonstrating how we can discover history in news archives. In this demo we focus on demonstrating two of the end user tools available in the prototype, the cross-discipline search and the Papyrus browser.
This paper describes the 'LipTelephone' system, which is being developed under a Greek AM... more This paper describes the 'LipTelephone' system, which is being developed under a Greek AMEA research project, with the purpose to serve as a videophone that can also be used by lip readers. This system aims to combine model-based with traditional coding techniques in order to exploit the information redundancy in a scene of known content, while achieving high fidelity representation in the specific area of interest, which is the speaker's mouth. The recent MPEG-4 coding standard provides the framework, which is employed, for these purposes.
Proceedings of the Conference for Truth and Trust Online 2019
This paper presents an overview of the WeVerify H2020 EU project, which is developing intelligent... more This paper presents an overview of the WeVerify H2020 EU project, which is developing intelligent human-in-the-loop content verification and disinformation analysis methods. Social media and web content are analysed and contextualised within the broader online ecosystem, in order to expose fabricated content, through cross-modal content verification, social network analysis, microtargeted debunking and a blockchain-based public database of known fakes. In particular, we introduce the following already developed tools: cross-modal content verification; blockchain-based database of known fakes; an open source content verification browser plugin; a collaborative verification workbench.
Proceedings of the 23rd International Conference on World Wide Web - WWW '14 Companion, 2014
ABSTRACT There are both positive and negative aspects in the use of social media in news and info... more ABSTRACT There are both positive and negative aspects in the use of social media in news and information dissemination. To deal with the negative aspects, such as the spread of rumours and fake news, the flow of information should implicitly be filtered and marked to specific criteria such as credibility, trustworthiness, reputation, popularity, influence, and authenticity. This paper proposes an approach that can enhance trustworthiness and content validity in the presence of information overload. We introduce Alethiometer, a framework for assessing truthfulness in social media that can be used by professional and general news users alike. We present different measures that delve into the detailed analysis of the content, the contributors of the content and the underlying context. We further propose an approach for deriving a single metric that considers, in a unified manner, the quality of a contributor and of the content provided by that contributor. Finally, we present some preliminary statistical results from the examination of a set of 10 million twitter users, that provide useful insights on the characteristics of social media data.
Keywords: 3D model based coding, rigid head tracking, 3D motion estimation, feature tracking The ... more Keywords: 3D model based coding, rigid head tracking, 3D motion estimation, feature tracking The presented work proposes substantial improvements to established methods for the twodimensional tracking of an image and the rigid adaptation of a three-dimensional face model to a 2D projection. These are combined to produce a robust system for the rigid 3D tracking of a human head which is promising even for extreme rotation conditions where most methods fail. Our technique, while using a 2D feature-based image tracking algorithm introduced by Kanade, Lucas and Tomasi, is based on the assessment of the tracked features considering their suitability for tracking, the success in tracking them in the current frame and their reliability judged from their 3D position in a model adapted to the previous frame. Having determined the optimum set of tracked features on the current image frame these are used to readapt the 3D human face model which has been initialised to fit to the face manually ...
There are both positive and negative aspects in the use of social media in news and information d... more There are both positive and negative aspects in the use of social media in news and information dissemination. To deal with the negative aspects, such as the spread of rumours and fake news, the flow of information should implicitly be filtered and marked to specific criteria such as credibility, trustworthiness, reputation, popularity, influence, and authenticity. This paper proposes an approach that can enhance trustworthiness and content validity in the presence of information overload. We introduce Alethiometer, a framework for assessing truthfulness in social media that can be used by professional and general news users alike. We present different measures that delve into the detailed analysis of the content, the contributors of the content and the underlying context. We further propose an approach for deriving a single metric that considers, in a unified manner, the quality of a contributor and of the content provided by that contributor. Finally, we present some preliminary s...
The tourism & leisure-related information contents provided by the services identified at the ASK... more The tourism & leisure-related information contents provided by the services identified at the ASK-IT pilot sites have about the same quality as the examined site-independent best-practice examples. The observations reported in the previous section are hence also valid for the siterelated services: The services focus on specific user groups while neglecting others. The service contents show generally also the same bias toward user groups with physical impairments.
� Abstract— This is a brief description of MESH - Multimedia semantic syndication for enhanced ne... more � Abstract— This is a brief description of MESH - Multimedia semantic syndication for enhanced news services (http://www.mesh-ip.eu), an EC co-funded project that started in March 2006 to create a framework for intelligent creation and delivery of semantically-enhanced multimedia news.
International Workshop on Intelligent Communication Technologies and Applications with Emphasis on Mobile Communications, May 5, 1999
The presented work proposes substantial improvements to established methods for the twodimensiona... more The presented work proposes substantial improvements to established methods for the twodimensional tracking of an image and the rigid adaptation of a three-dimensional face model to a 2D projection. These are combined to produce a robust system for the rigid 3D tracking of a human head which is promising even for extreme rotation conditions where most methods fail. Our technique, while using a 2D feature-based image tracking algorithm introduced by Kanade, Lucas and Tomasi, is based on the assessment of the tracked ...
Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101), 2000
This paper addresses the problem of adapting a generic 3D face model to a human face of which the... more This paper addresses the problem of adapting a generic 3D face model to a human face of which the frontal and profile views are given. Assuming that a set of feature points have been detected on both views the adaptation procedure initializes with a rigid transformation of the model aiming to minimize the distances of the 3D model feature nodes from the calculated 3D coordinates of the 2D feature points. Then, a non-rigid transformation ensures that the feature nodes are displaced optimally close to their exact calculated positions, dragging their neighbors in a way that does not deform the facial model in an unnatural way.
Technology, Management and Applications, 2002
Proceedings of the 5th Seminar on Neural Network Applications in Electrical Engineering. NEUREL 2000 (IEEE Cat. No.00EX287), 2000
Proceedings of International Conference on Image Processing, 1997
A coding method for videophone communications over wireless, varying bit-rate, channels, is propo... more A coding method for videophone communications over wireless, varying bit-rate, channels, is proposed. This method is based on defining one or more areas of interest and breaking them into a number of arbitrarily shaped regions which are homogeneous in their motion. The interest areas in the current implementation are human faces. They are detected with the help of a 2D
Lecture Notes in Computer Science, 2011
Digital archives comprise a valuable asset for effective information retrieval. In many cases, ho... more Digital archives comprise a valuable asset for effective information retrieval. In many cases, however, the special vocabulary of the archive restricts its access only to experts in the domain of the material it contains and, as a result, researchers of other disciplines or the general public cannot take full advantage of the wealth of information it offers. To this end, the Papyrus research project has worked towards a solution which makes cross-discipline search possible in digital libraries. The developed prototype showcases this approach demonstrating how we can discover history in news archives. In this demo we focus on demonstrating two of the end user tools available in the prototype, the cross-discipline search and the Papyrus browser.