Search

This paper presents a basic configuration of a system that provides dynamic delivery of full-motion video while following target users in ubiquitous computing environments. The proposed system is composed of multiple computer displays with radio frequency identification (RFID) tag readers, which are automatically connected to a network via IP, and RFID tags worn by users and some network servers. We adopted a passive tag RFID system. The delivery of full-motion video uses adaptive broadcasting. The system can continuously deliver streaming data, such as full-motion video, to the display, through the database and the streaming server on the network, moving from one display to the next as the user moves through the network. Because it maintains the information about the user's location in real time, it supports the user wherever he or she is, without requiring a conscious request to obtain their information. This paper describes a prototype implementation of this framework and a practical application. [ABSTRACT FROM AUTHOR]

This paper describes two user-centered studies undertaken in the development of a concurrent group voice and media sharing application. The first used paper prototyping to identify the user values relating to a number of functional capabilities. These results informed the development of a prototype application, which was ported to a 3G handset and evaluated in the second study using a conjoint analysis approach. Results indicated that concurrent photo sharing was of high user value, while the value of video sharing was limited by established mental models of file sharing. Overall higher ratings were found among female subjects and among less technologically aware subjects and most media sharing would be with those who are close and trusted. This, and other results suggest that the reinforcement of social connections, spontaneity and emotional communications would be important user objectives of such a media sharing application. [ABSTRACT FROM AUTHOR]

With the proliferation of digital multimedia content, issues of copyright protection have become more important because the copying of digital video does not result in the decrease in quality that occurs when analog video is copied. One method of copyright protection is to embed a digital code, "watermark", into the video sequence. The watermark can then unambiguously identify the copyright holder of the video sequence. In this paper, we propose a new video watermarking algorithm for the H.264 coded video with considering usability factors. The usability testings based on the concept of Human Computer Interface (HCI) have been performed on the proposed approach. The usability testing has been considered representative for most image manipulations and attacks. The proposed algorithm has passed all the attack testings. Therefore, the watermarking mechanisms in this paper have been proved to be robust and efficient to protect the copyright of H.264 coded video. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new face and eye gaze tracking method that works by attaching gaze tracking devices to stereoscopic shutter glasses. This paper presents six advantages over previous works. First, through using the proposed method with stereoscopic VR systems, users feel more immersed and comfortable. Second, by capturing reflected eye images with a hot mirror, we were able to increase eye gaze accuracy in a vertical direction. Third, by attaching the infrared passing filter and using an IR illuminator, we were able to obtain robust gaze tracking performance irrespective of environmental lighting conditions. Fourth, we used a simple 2D-based eye gaze estimation method based on the detected pupil center and the ‘geometric transform' process. Fifth, to prevent gaze positions from being unintentionally moved by natural eye blinking, we discriminated between different kinds of eye blinking by measuring pupil sizes. This information was also used for button clicking or mode toggling. Sixth, the final gaze position was calculated by the vector summation of face and eye gaze positions and allowing for natural face and eye movements. Experimental results showed that the face and eye gaze estimation error was less than one degree. [ABSTRACT FROM AUTHOR]

This paper presented a gesture-based interaction system which provides a natural way of manipulating on-screen objects. We generate a synthetic image by linking images from two cameras to recognize hand gestures. The synthetic image contains all the features captured from two different views, which can be used to alleviate the self-occlusion problem and improve the recognition rate. The MDA and EM algorithms are used to obtain parameters for pattern classification. To compute more detailed pose parameters such as fingertip positions and hand contours in the image, a random sampling method is introduced in our system. We describe a method based on projective geometry for background subtraction to improve the system performance. Robustness of the system has been verified by extensive experiments with different user scenarios. The applications of picture browser and visual pilot are discussed in this paper. [ABSTRACT FROM AUTHOR]

In this paper, we propose gesture recognition in multiple people environment. Our system is divided into two modules: Segmentation and Recognition. In segmentation part, we extract foreground area from input image, and we decide the closest person as a recognition subject. In recognition part, firstly we extract feature point of subject's both hands using contour based method and skin based method. Extracted points are tracked using Kalman filter. We use trajectories of both hands for recognizing gesture. In this paper, we use the simple queue matching method as a recognition method. We also apply our system as an animation system. Our method can select subject effectively and recognize gesture in multiple people environment. Therefore, proposed method can be used for real world application such as home appliance and humanoid robot. [ABSTRACT FROM AUTHOR]

In this paper, we developed a fingertip finding algorithm working with a regular diffuser. The proposed algorithm works on images captured by infra-red cameras, settled on one side of the diffuser, observing human gestures taken place on the other side. With diffusion characteristics of the diffuser, we can separate finger-touch from palm-hover events when the user interacts with the diffuser. This paper contributes on: Firstly, the technique works with a regular diffuser, infra-red camera coupled with an infra-red illuminator, which is easy to deploy and cost effective. Secondly, the proposed algorithm is designed to be robust for casually illuminated surface. Lastly, with diffusion characteristics of the diffuser, we can detect finger-touch and palm-hover events, which is useful for natural user interface design. We have deployed the algorithm on a rear-projection multi-resolution tabletop, called I-M-Top. A video retrieval application using the two events on design of UIs is implemented to show its intuitiveness on the tabletop system. [ABSTRACT FROM AUTHOR]

The paper addresses a tour planning problem, which encompasses weakly specified constraints such as different kinds of activities together with corresponding spatial assignments such as locations and regions. Alternative temporal orders of planed activities together with underspecified spatial assignments available at different levels of granularity lead to a high computational complexity of the given tour planning problem. The paper introduces the results of an exploratory tour planning study and a Region-based Direction Heuristic, derived from the acquired data. A gesture-based interaction model is proposed, which allows structuring the search space by a human user at a high level of abstraction for the subsequent generation of alternative solutions so that the proposed Region-based Direction Heuristic can be applied. [ABSTRACT FROM AUTHOR]

This paper describes ongoing work towards building a multimodal computer system capable of sensing the affective state of a user. Two major problem areas exist in the affective communication research. Firstly, affective states are defined and described in an inconsistent way. Secondly, the type of training data commonly used gives an oversimplified picture of affective expression. Most studies ignore the dynamic, versatile and personalised nature of affective expression and the influence that social setting, context and culture have on its rules of display. We present a novel approach to affective sensing, using a generic model of affective communication and a set of ontologies to assist in the analysis of concepts and to enhance the recognition process. Whilst the scope of the ontology provides for a full range of multimodal sensing, this paper focuses on spoken language and facial expressions as examples. [ABSTRACT FROM AUTHOR]

We present our work on emotionally expressive avatars, animated virtual characters that can express emotions via facial expressions. Because these avatars are highly distinctive and easily recognizable, they may be used in a range of applications. In the first part of the paper we present their use in computer mediated communication where two or more people meet in virtual space, each represented by an avatar. Study results suggest that social interaction behavior from the real-world is readily transferred to the virtual world. Empathy is identified as a key component for creating a more enjoyable experience and greater harmony between users. In the second part of the paper we discuss the use of avatars as an assistive, educational and therapeutic technology for people with autism. Based on the results of a preliminary study, we provide pointers regarding how people with autism may overcome some of the limitations that characterize their condition. [ABSTRACT FROM AUTHOR]

This paper deals with multi-channel interaction managing thru an intermediation between channels and Interactive e-Services (IeS). After work on modeling and theoretical framework, we implemented a platform: Ubi-Learn, which is able to manage this kind of interaction thru an intermediation middleware based on a Multi-Agents System (MAS): Jade. The issue addressed here is linked to the way you choose a channel depending on the user's task. First, we have encoded several ad hoc rules (tacit knowledge) into the system. In this paper, we present our new approach based on association rules mining approach which allows us to propose automatically several dynamic rules (explicit knowledge). [ABSTRACT FROM AUTHOR]

This paper describes an e-learning interface with multiple tutoring character agents. The character agents use eye movement information to facilitate empathy-relevant reasoning and behavior. Eye Information is used to monitor user's attention and interests, to personalize the agent behaviors, and for exchanging information of different learners. The system reacts to multiple users' eye information in real-time and the empathic character agents owned by each learner exchange learner's information to help to form the online learning community. Based on these measures, the interface infers the focus of attention of the learner and responds accordingly with affective and instructional behaviors. The paper will also report on some preliminary usability test results concerning how users respond to the empathic functions and interact with other learners using the character agents. [ABSTRACT FROM AUTHOR]

In software design, the reuse issue brings the increasing of web services, components and others techniques. These techniques allow reusing code associated to technical aspect (as software component). With the development of business components which can integrate technical aspect with HCI, the composition issue has appeared. Our previous work concerned the GUI composition based on an UIDL as UsiXML. With the generalization of Multimodal User Interfaces (MUI), MUI composition principles have to be studied. This paper aims at extend existing basic composition principles in order to treat multimodal interfaces. The same principle as in the previous work, based on the tree algebra, can be used in another level (AUI) of the UsiXML framework to support the Multimodal User Interfaces composing. This paper presents a case study on the food ordering system based on multimodal (coupling GUI and MUI). A conclusion and the future works in the HCI domain are presented. [ABSTRACT FROM AUTHOR]

Voice packets with guaranteed QoS (Quality of Service) on the VoIP system are responsible for digitizing, encoding, decoding, and playing out the speech signal. The important point is based on the factor that different parts of speech over IP networks have different perceptual importance and each part of speech does not contribute equally to the overall voice quality. In this paper, we propose new additive noise reduction algorithms to improve voice over IP networks and present performance evaluation of perceptual speech signal through IP networks in the additive noise environment during realtime phone-call service. The proposed noise reduction algorithm is applied to pre-processing method before speech coding and to post-processing method after speech decoding based on single microphone VoIP system. For noise reduction, this paper proposes a Wiener filter optimized to the estimated SNR of noisy speech for speech enhancement. Various noisy conditions including white Gaussian, office, babble, and car noises are considered with G.711 codec. Also, we provide critical message report procedures and management schemes to guarantee QoS over IP networks. Finally, as following the experimental results, the proposed algorithm and method has been prove for improving speech quality. [ABSTRACT FROM AUTHOR]

The word boundary detection has an application in speech processing systems. The problem this paper tries to solve is to separate words of a sequence of phonemes where there is no delimiter between phonemes. In this paper, at first, a recurrent fuzzy neural network (RFNN) together with its relevant structure is proposed and learning algorithm is presented. Next, this RFNN is used to predict word boundaries. Some experiments have already been implemented to determine complete structure of RFNN. Here in this paper, three methods are proposed to encode input phoneme and their performance have been evaluated. Some experiments have been conducted to determine required number of fuzzy rules and then performance of RFNN in predicting word boundaries is tested. Experimental results show an acceptable performance. [ABSTRACT FROM AUTHOR]

The goal of our research has been to investigate the different ways in with using new communication technologies, especially mobile multimedia communications, could support the city-based communities. In this paper we review the research done about the effect of mobile technology, specially mobile video, into communities' communication patterns, and highlight the new challenges and gaps still not covered in this area. Finally, we will describe how we have tried to respond to these challenges by using User Centered Design in two very different types of communities: women associations, and elderly people. [ABSTRACT FROM AUTHOR]

In this paper, a place model based on a feature distribution is proposed for place recognition. In many previous proposed methods, places are modeled as images or a set of extracted features. In those methods, a database of images or feature sets should be built. The cost of search time will grow exponentially when the database goes large. The proposed feature distribution method uses global information of each place and the search space grows linearly according to the number of places. In the experiments, we evaluate the performance using different number of frames and features for the recognition each time. Additionally, we have shown that the proposed method is applicable to many real-time applications such as robot navigation, wearable computing systems, and so on. [ABSTRACT FROM AUTHOR]

A smart home is an intelligent and shared space, where various services coexist and multiple residents with different preferences and habits share these services most of the time. Due to the sharing of space and time, service conflicts may occur when multiple users try to access media services. In this paper, we propose a context-based mediation method, consisting of service mediators and mobile mediators, to resolve the service conflicts in a smart home. The service mediators detect service conflicts among the residents and recommend their preferred media contents on a shared screen and their own mobile devices by exploiting users' preferences and service profiles. The mobile mediators collect the recommendation information and give the users personal recommendation. With combination of the service and mobile mediator, the residents are allowed to negotiate the media contents in the conflict situation. Based on experiments in the ubiHome, we observed that mediation is useful to encourage discussion and helps to choose a proper service in a conflict situation. Therefore, we expect the proposed mediation method to play a vital role in resolving conflicts and providing multiple residents with harmonized services in a smart home environment. [ABSTRACT FROM AUTHOR]

The ease-of-use of Web-based video-publishing services provided by applications like YouTube has encouraged a new means of asynchronous communication, in which users can post videos not only to make them public for review and criticism, but also as a way to express moods, feelings, or intentions to an ever-growing network of friends. Following the current trend of porting Web applications onto mobile platforms, the authors sought to explore user-interface design issues of a mobile-device-based YouTube, which they call m-YouTube. They first analyzed the elements of success of the current YouTube Web site and observed its functionality. Then, they looked for unsolved issues that could give benefit through information-visualization design for small screens on mobile phones to explore a mobile version of such a product/service. The biggest challenge was to reduce the number of functions and amount information to fit into a mobile phone screen, but still be usable, useful, and appealing within the YouTube context of use and user experience. Borrowing ideas from social research in the area of social influence processes, they made design decisions aiming to help YouTube users to make the decision of what video content to watch and to increase the chances of YouTube authors being evaluated and observed by peers. The paper proposes a means to visualize large amounts of video relevant to YouTube users by using their friendship network as a relevance indicator to help in the decision-making process. [ABSTRACT FROM AUTHOR]

In this paper we describe a new interface for browsing and sorting of digital pictures. Our approach is two-fold. First we present a new method to automatically identify similar images and rate them based on sharpness and exposure quality of the images. Second we present a zoomable user interface based on the details-on-demand paradigm enabling users to browse large collections of digital images and select only the best images for further processing or sharing. [ABSTRACT FROM AUTHOR]

With rapid growth of the mobile industry, the limitation of small screen mobile is attracting a lot of researchers attention for transforming on/off-line contents into mobile contents. Frame segmentation for limited mobile browsers is the key point of off-line contents tranformation. The X-Y recursive cut algorithm has been widely used for frame segmentation in document analysis. However, this algorithm has drawbacks for cartoon images which have various image types and image with noises, especially the online cartoon contents obtain during scanning. In this paper, we propose a method to segment on/off-line cartoon contents into fitted frames for the mobile screen. This makes the x-y recursive cut algorithm difficult to find the exact cutting point. Therefore we use a method by combining two concepts: an X-Y recursive cut algorithm to extract candidate segmenting positions which shows a good performance on noises free contents, and Multi-Layer Perceptrons (MLP) concept use on candidate for verification. These methods can increase the accuracy of the frame segmentation and feasible to apply on various off-line cartoon images with frames. [ABSTRACT FROM AUTHOR]

In this paper we present G-tunes, a music player that couples tangible interface with digital music. The design is done based on the research of tangible interface and interaction engineering. We offer an overview of design concept, explain the prototyping and discuss the result. One of the goals of this project is to create rich experiences for people to play music; another goal is to explore how external physical expressions relate to human's inner perception and emotion, and how we can couple this with the design of a tangible music player. [ABSTRACT FROM AUTHOR]

We present en experimental evaluation of VISTO (Vector Image Search TOol), a new content-based image retrieval (CBIR) system that deals with vector images in SVG (Scalable Vector Graphics) format, differently to most of the CBIR tools available in the literature that deal with raster images. The experimental evaluation of retrieval systems is a critical part in the process of continuously improving the existing retrieval metrics. While researchers in text image retrieval have long been using a sophisticated set of tools for user-based evaluation, this does not yet apply to image retrieval. In this paper, we make a step forward toward this direction and present an experimental evaluation of VISTO in a framework for the production of 2D animation. [ABSTRACT FROM AUTHOR]

We can classify various images as either people pictures, if they contain one or more persons, or scenery pictures, if they lack people, by using face region detection. However, the precision of a picture's classification is low if it uses existing face region detection technique. This paper proposes the algorithm about structured simplicity of the picture to do classification with higher accuracy. To verify the usefulness of an offer method, we did a classification experiment which uses 500 people pictures and scenery pictures. The experiment to use only face region detection in Open CV showed an accuracy of 79% detection rate. While the experiment to use face region detection in structured simplicity with Open CV showed an accuracy of 86.4%. Therefore by using structured simplicity with face region detection, we can do an efficient picture classification of a person picture and scenery picture. [ABSTRACT FROM AUTHOR]

This paper proposes a new Error Concealment (EC) method for the H.264/AVC [1] video coding standard using both spatial and temporal information for intra-frame concealment. Five error concealing modes are offered by this method. The proposed EC method also allows feedback from users. It allows users to define and change the thresholds for switching between five different modes during the error concealing procedure. As a result, the concealing result for a video sequence can be optimized by taking advantage of relevant user feedback. The concealed video quality has been measured by a group of users and compared with the H.264 EC method which is without user feedback. The experimental results show that the proposed new EC algorithm with the user feedback performs better (3 dB gains) than the H.264 EC without user feedback. [ABSTRACT FROM AUTHOR]

This paper aims to explore the quality of the user experience with mobile and pervasive interactive multimedia systems that enable the creation and sharing of digital content through mobile phones. It also looks at discussing the use and validity of different experimental in-situ and other data gathering and evaluation techniques for the assessment of how the physical and social contexts might influence the use of these systems. This scenario represents an important shift away from professionally produced digital content for the mass-market. It addresses methodologies and techniques that are suitable to design co-creative applications for non-professional users in different contexts of use at home or in public spaces. Special focus is be given to understand how user participation and motivation in small themed communities can be encouraged, and how social interaction can be enabled through mobile interfaces. An enhancement of users creativity, self-authored content sharing, sociability and co-experience can be evidence for how creative people can benefit from Information and Communication Technologies. [ABSTRACT FROM AUTHOR]

The ISO 9241-9 standard for computer pointing devices proposes an evaluation of performance and comfort [4]. This paper is the first eye tracking evaluation conforming to ISO 9241-9. We evaluated three techniques and compared them with a standard mouse. The evaluation used throughput (in bits/s) as a measurement of user performance in a multi-directional point-select task. The "Eye Tracking Long" technique required participants to look at an on-screen target and dwell on it for 750 ms for selection. Results revealed a lower throughput than for the "Eye Tracking Short" technique with a 500 ms dwell time. The "Eye+Spacebar" technique allowed participants to "point" with the eye and "select" by pressing the spacebar upon fixation. This eliminated the need to wait for selection. It was the best among the three eye tracking techniques with a throughput of 3.78 bits/s, which was close to the 4.68 bits/s for the mouse. [ABSTRACT FROM AUTHOR]

This paper presents a novel Human-Computer Interaction (HCI) system with calibrated mono-camera which integrates active computer vision technology and embedded speech command recognition technology. Mainly by tracking the nose tip motion robustly as the mouse trace, this system completes mouse mission with recognition rate more than 85% at the speed 15 frame per second. To achieve the goal, we adopt a novel approach based on the symmetry of the nose plane feature to localize and track invariantly to the varied environment. Comparing to other kinds of pointing device, this hand-free HCI system is hands-free, cheap, real-time, convenient and unpolluted, which can be used in the field of disabled aid, entertainment and remote control. [ABSTRACT FROM AUTHOR]

In this paper, we propose a system for multiple people tracking using fragment based histogram matching. Appearance model is based on Improved HLS color histogram which can be calculated efficiently using integral histogram representation. Since the histograms will loss all spatial information, we define a fragment based region representation which retains spatial information, robust against occlusion and scale issue by using disparity information. Multiple people labeling is maintained by creating an online appearance representation for each person detected in the scene and calculating fragment vote map. Initialization is performed automatically from the background segmentation step. [ABSTRACT FROM AUTHOR]

The Gesture and Activity Recognition Toolit (GART) is a user interface toolkit designed to enable the development of gesture-based applications. GART provides an abstraction to machine learning algorithms suitable for modeling and recognizing different types of gestures. The toolkit also provides support for the data collection and the training process. In this paper, we present GART and its machine learning abstractions. Furthermore, we detail the components of the toolkit and present two example gesture recognition applications. [ABSTRACT FROM AUTHOR]

In this paper, we present a human shape extraction and tracking for gait recognition using geodesic active contour models(GACMs) combined with mean-shift algorithm. The active contour models (ACMs) are very effective to deal with the non-rigid object because of its elastic property, but they have the limitation that their performance is mainly dependent on the initial curve. To overcome this problem, we combine the mean-shift algorithm with the traditional GACMs. The main idea is very simple. Before evolving using level-set method, the initial curve in each frame is re-localized near the human region and is resized enough to include the targe object. This mechanism allows for reducing the number of iterations and for handling the large object motion. Our system is composed of human region detection and human shape tracking. In the human region detection module, the silhouette of a walking person is extracted by background subtraction and morphologic operation. Then human shape are correctly obtained by the GACMs with mean-shift algorithm. To evaluate the effectiveness of the proposed method, it is applied the common gait data, then the results show that the proposed method is extracted and tracked efficiently accurate shape for gait recognition. [ABSTRACT FROM AUTHOR]

In this paper, we design an Attention Focus Kalman Filter (AFKF) - a framework that offers interaction capabilities by constructing an eye-movement language, provides real-time perceptual compression through Human Visual System (HVS) modeling, and improves system's reliability. These goals are achieved by an AFKF through identification of basic eye-movement types in real-time, the prediction of a user's perceptual attention focus, and the use of the eye's visual sensitivity function and eye-position data signal de-noising. [ABSTRACT FROM AUTHOR]

In this paper, we propose a real-time face tracking system using adaptive face detector and the Kalman filter. Basically, the features used for face detection are five types of simple Haar-like features. To only extract the more significant features from these features, we employ principal component analysis (PCA). The extracted features are used for a learning vector of the support vector machine (SVM), which classifies the faces and non-faces. The face detector locates faces from the face candidates separated from the background by using real-time updated skin color information. We trace the moving faces with the Kalman filter, which uses the static information of the detected faces and the dynamic information of changes between previous and current frames. In this experiment, the proposed system showed an average tracking rate of 97.3% and a frame rate of 23.5 frames per s, which can be adapted into a real-time tracking system. [ABSTRACT FROM AUTHOR]

In this paper, we propose a gesture modeling system based on computer vision in order to recognize a gesture naturally without any trouble between a system and a user using real-time 3D modeling information on multiple objects. It recognizes a gesture after 3D modeling and analyzing the information pertaining to the user's body shape in stereo views for human movement. In the 3D-modeling step, 2D information is extracted from each view by using an adaptive color difference detector. Potential objects such as faces, hands, and feet are labeled by using the information from 2D detection. We identify reliable objects by comparing the similarities of the potential objects that are obtained from both the views. We acquire information on 2D tracking from the selected objects by using the Kalman filter and reconstruct it as a 3D gesture. A joint of each part of a body is generated in the combined objects. We experimented on ambiguities using occlusion, clutter, and irregular 3D gestures to analyze the efficiency of the proposed system. In this experiment, the proposed gesture modeling system showed a good detection and a processing time of 30 frames per second, which can be used in a real-time. [ABSTRACT FROM AUTHOR]

In this paper, we propose an approach toward body parts representation, localization, and human pose estimation from an image. In the image, the human body parts and a background are represented by a mixture of Gaussians, and the body parts configuration is modeled by a Bayesian network. In this model, state nodes represent pose parameters of an each body part, and arcs represent spatial constraints. The Gaussian mixture distribution is used to model the prior distribution for the body parts and the background as a parametric model. We estimate the human pose through an optimization of the pose parameters using likelihood objective functions. The performance of the proposed approach is illustrated on various single images, and improves the human pose estimation quality. [ABSTRACT FROM AUTHOR]

This paper presents a novel computer game system that controls a game using only the movement of human's facial features. Our system is specially designated for the handicapped people with severe disabilities and the people without experience of using the computer. Using a usual PC camera, the proposed game system detects the user's eye movement and mouse movement, and then interprets the communication intent to play a game. The game system is tested with 42 numbers of people, and then the result shows that our game system should be efficiently and effectively used as the interface for the disabled people. [ABSTRACT FROM AUTHOR]

A prerequisite component of a autonomous mobile vehicle system is the self localization ability to recognize its environment and to estimate where it is. Generally, we can determine the position and the pose using homography approach, but it has errors especially in simultaneous change of position and pose. In this paper, we proposed position and pose computation method of a camera through analysis of images obtained from camera equipped mobile robot. Proposed method is made up of two steps. First step is to extract feature points and matching in sequential images. Second step is to compute the accurate camera position and pose using geometric edge matching. In first step, we use KLT tracking to extract feature points and matching in sequential images. In second step, we propose an iterative matching method between predicted edge models through perspective transform using the result calculated by homography of the matched feature points and generated edge models in correspond points till there is no variation in matching error. For the purpose of the performance evaluation, we performed the test to compensate the position and the pose of the camera installed in wireless-controlled vehicle with the video sequence stream obtained at 15Hz frame rate and show the experimental results. [ABSTRACT FROM AUTHOR]

In this paper we present our experience in using optical tracking systems in Virtual Environment applications. First we briefly describe the tracking systems we used, and then we describe the application scenarios and present how we adapted the scenarios for the tracking systems. One of the tracking systems is markerless, that means that a user doesn't have to wear any specific devices to be tracked and can interact with an application with free hand movements. With our application we compare the performance of different tracking systems and demonstrate that it is possible to perform complex actions in an intuitive way with just small special knowledge of the system and without any specific devices. This is a step forward to a more natural human-computer interface. [ABSTRACT FROM AUTHOR]

The wide availability of digital reading material online is causing a major shift in everyday reading activities. Readers are skimming instead of reading in depth [Nielson 1997]. Highlights are increasingly used in digital interfaces to direct attention toward relevant passages within texts. In this paper, we study the eye-gaze behavior of subjects using both keyword highlighting and ScentHighlights [Chi et al. 2005]. In this first eye-tracking study of highlighting interfaces, we show that there is direct evidence of the von Restorff isolation effect [VonRestorff 1933] in the eye-tracking data, in that subjects focused on highlighted areas when highlighting cues are present. The results point to future design possibilities in highlighting interfaces. [ABSTRACT FROM AUTHOR]

This paper describes an approach for a system which analyses an orchestra conductor in real-time, with the purpose of using the extracted information of time pace and expression for an automatic play of a computer-controlled instrument (synthesizer). The system in its final stage will use non-intrusive computer vision methods to track the hands of the conductor. The main challenge is to interpret the motion of the hand/baton/mouse as beats for the timeline. The current implementation uses mouse motion to simulate the movement of the baton. It allows to "conduct" a pre-stored MIDI file of a classical orchestral music work on a PC. [ABSTRACT FROM AUTHOR]

This paper proposes an novel two-pass adaptation method for online language identification by using confidence measure based incremental language model adaptation. In this system, we firstly used semi-supervised language model adaptation to solve the problem of channel mismatch, and then used unsupervised incremental adaptation to adjust new language model during online language identification. For robust adaptation, we compare three confidence measures and then present a new fusion method with Bayesian classifier. Tested on the RMTS(Real-world Multi-channel Telephone Speech) database, experiments show that using semi-supervised language model adaptation, the target language detection rate rises from 73.26% to 80.02% and after unsupervised incremental language model adaptation, an extra rise over 3.91% (from 80.02% to 83.93%) is obtained. [ABSTRACT FROM AUTHOR]

Aiming at emotion deficiency in present E-Learning system, speech emotion recognition system is proposed in the paper. A corpus of emotional speech from various subjects, speaking different languages is collected for developing and testing the feasibility of the system. The potential prosodic features are first identified and extracted from the speech data. Then we introduce a systematic feature selection approach which involves the application of Sequential Forward Selection (SFS) with a General Regression Neural Network (GRNN) in conjunction with a consistency-based selection method. The selected features are employed as the input to a Modular Neural Network (MNN) to realize the classification of emotions. Our simulation experiment results show that the proposed system gives high recognition performance. [ABSTRACT FROM AUTHOR]

This paper describes the aspects kept in mind for the development of the tutoring and domain models, of an Intelligent Tutoring System (ITS), where the instruction type that will give the tutoring system, the pedagogic strategies and the structure of the course are established. Also is described the software development process and their principal functions. This work is part of the research project that involves the adaptation process of the interfaces into Intelligent Tutoring Systems at the University of Sinu's TESEEO Research Group ([2]). The final objective of this work is to provide mechanisms for the design and development of system interfaces for tutoring/training, those are effective and at the same time modular, structured, configurable, flexible and adaptable. [ABSTRACT FROM AUTHOR]

This paper focuses on a method to extract video browsing styles and reusing it. In video browsing process for knowledge work, users often develop their own browsing styles to explore the videos because the domain knowledge of contents is not enough, and then the users interact with videos according to their browsing style. The User Experience Reproducer enables users to browse new videos according to their own browsing style or other users' browsing styles. The preliminary user studies show that video browsing styles can be reused to other videos. [ABSTRACT FROM AUTHOR]

Predicting user behavior is an important issue in Human Computer Interaction ([5]) research, having an essential role when developing intelligent user interfaces. A possible solution to deal with this challenge is to build an intelligent interface agent ([8]) that learns to identify patterns in users behavior. The aim of this paper is to introduce a new agent based approach in predicting users behavior, using a probabilistic model. We propose an intelligent interface agent that uses a supervised learning technique in order to achieve the desired goal. We have used Aspect Oriented Programming ([7]) in the development of the agent in order to benefit of the advantages of this paradigm. Based on a newly defined evaluation measure, we have determined the accuracy of the agent's prediction on a case study. [ABSTRACT FROM AUTHOR]

Complex systems are characterized by a large number and variety of, and often a high degree of dependency between, subsystems. Complexity, in combination with coupling, has been shown to lead to difficulties with monitoring and comprehending system status and activities and thus to an increased risk of breakdowns in human-machine coordination. In part, these breakdowns can be explained by the fact that increased complexity tends to be paralleled by an increase in the amount of data that is made available to operators. Presenting this data in an inappropriate form is crucial to avoiding problems with data overload and attention management. One approach for addressing this challenge is to move from fixed display designs to adaptive information presentation, i.e., information presentation that changes as a function of context. This paper will discuss possible approaches to, challenges for, and effects of increasing the flexibility of information presentation. [ABSTRACT FROM AUTHOR]

The effective retrieval and multimedia data management techniques to facilitate the searching and querying of large multimedia data sets are very important in multimedia applications development. The content-based retrieval systems must use the multimedia content to represent and to index data. The representation of multimedia data supposes to identify the most useful features for representing the multimedia content and the approaches needed for coding the attributes of multimedia data. The multimedia content adaptation realize the multimedia resources manipulation, respecting the specific quality parameters, function on the limits required by networks and terminal devices. The goal of the paper is to identify a design model for using content-based multimedia data retrieval in multimedia content adaptation. The goal of this design is to deliver the multimedia content in various networks and to different types of peripheral devices, in the most appropriate format and function on specific characteristics. [ABSTRACT FROM AUTHOR]

Interest in emotion detection is increasing significantly. For research and development in the field of Affective Computing, in smart environments, but also for reliable non-lab medical and psychological studies or human performance monitoring, robust technologies are needed for detecting evidence of emotions in persons under everyday conditions. This paper reports on evaluation studies of the EREC-II sensor system for acquisition of emotion-related physiological parameters. The system has been developed with a focus on easy handling, robustness, and reliability. Two sets of studies have been performed covering 4 different application fields: medical, human performance in sports, driver assistance, and multimodal affect sensing. Results show that the different application fields pose different requirements mainly on the user interface, while the hardware for sensing and processing the data proved to be in an acceptable state for use in different research domains. [ABSTRACT FROM AUTHOR]

This paper proposed a semantic over-viewing method. This method extracts headings and semantic blocks by analyzing a layout structure of a web page and can provide a semantic overview of the web page. This method allows users grasp the overall structure of pages. It also reduces the number of operations to target information to about 6% by moving along semantic blocks. Additionally, it reduces the cost of Web page creation because of adapting one Web page content to multi-terminals. The evaluations were conducted in respect to effectiveness, efficiency and satisfaction. The results confirmed that the proposed browser is more usable than the traditional method. [ABSTRACT FROM AUTHOR]

Collaborative filtering is a social information recommendation/ filtering method, and the peer-to-peer (P2P) computer network is a network on which information is distributed on the peer-to-peer basis (each peer node works as a server, a client, and even a router). This research aims to develop a model of P2P information recommendation system based on collaborative filtering and evaluate the ability of the system by computer simulations based on the model. We previously proposed a simple model, and the model in this paper is a modified one that is more focused on recommendation agents and user-agent interactions. We have developed a computer simulator program and tested simulations with several parameter settings. From the results of the simulations, recommendation recall and precision are evaluated. Findings are that the agents are likely to overly recommend so that the recall score becomes high but the precision score becomes low. [ABSTRACT FROM AUTHOR]