JMM Abstracts 

Vol.1 No.2 June, 2005

Editorial (pp091-091)
        D. Taniar   

Research articles:
Gesture Interaction for Small Handheld Devices to Support Multimedia Applications (pp092-111)
        J. Mäntyjärvi, S. Kallio, P. Korpipää, J. Kela and J. Plomp  
Accelerometer-based gesture control is proposed as a complementary interaction modality for small handheld devices to enable a variety of multimedia applications. The motivation for experimenting with gesture interaction is justified by the personal and public domain prototype applications developed. The challenges related to developing user-dependent and independent gesture control are presented. In this article, we experiment with methods for user-dependent gesture recognition with a low number of training repetitions, and for feasible user-independent gesture recognition from a moderately large set of gestures. The user-dependent gesture recognition performance of the continuous Hidden Markov Model (HMM) is better when compared to discrete HMM with three gesture repetitions in a training set. With continuous HMM, a recognition accuracy level of 95% is obtained with or without tilt normalization, while for discrete HMM a best recognition accuracy of 90% is obtained. The user-independent gesture recognition performance with continuous HMM of 89% is considerably better compared to tests with discrete HMM, when both are obtained with cross-validation from 2,520 gestures. An important result is that the effect of using tilt normalization notably increases the user-independent gesture recognition performance by 10-15% depending on the method used. The chosen methods show great potential for gesture-based interaction in multimedia applications.

From Motion to Emotion: A Wearable System for the Multimedia Enrichment of A Butoh Dace Performance (pp112-132)
        M. Barry, J. Gutknecht, I. Kulka, P. Lukowicz and T. Stricker
We present a mobile, multimedia system based on a network of body worn motion sensors, a wearable computer and a visualization engine that is used to produce a visual enhancement of Butoh dance performance. The core of the system is a novel motion classification scheme that allows us to capture the emotion expressed by the dancer during the performance and map it onto scripted visual effects. We describe the artistic concept behind the multimedia enhancement, the motion classification scheme and the system architecture. In an experimental evaluation we investigate the usefulness and the robustness of the wearable computer as well as the classification accuracy of the motion-sensing system. We also summarize the experiences with using the system for live performances on stage in several shows.

Human Face Animation Based on Video Analysis, with Applications to Mobile Entertainment (pp133-148)
        J. S.-S. Tang, A. W.-C. Liew and H. Yan
In this paper, we present a novel non-uniform rational B-spline (NURBS) muscle system for the simulation of human facial expressions and talking animation based on features extracted from video sequences. We construct the facial muscles based on anatomical knowledge and NURBS models. We use 43 feature points to represent facial expressions and apply a lip contour extraction technique to determine lip shapes. Our system is able to generate many different facial expressions and mouth shapes. The system has been used in web and mobile phone based digital entertainment applications.

Efficient Image Indexing and Retrieval over Mobile Devices (pp149-160)
        S. So  
The multimedia capabilities of mobile phones are rapidly increasing in recent years. It is now common that users can send and retrieve images over mobile phones. Image searching and retrieval is fundamentally becoming an important operation and yet there isní»t any efficient and effective way to do this over mobile phones.  The situation is particularly more acute when users attempt to access a content provider with a large collection of image contents.  Any aimless browsing of the images will translate into air-time costs. Therefore, an efficient and effective mechanism to retrieve images over mobile phones must be sought.      In this paper, we attempt to advance this area of mobile multimedia by providing a data model as well as a query model for searching images efficiently and effectively. The data model is a semantically rich structure for representing any salient features in the image contents. XML Schema is used to model the data structure and a number of examples are illustrated in this paper. The query model works intimately with the data model and it goes beyond simple Boolean type queries. The query model supports ranked queries from the repository of image contents. The most precisely matched images will be delivered first. Users can browse the ranked images through the logical expression of the query over mobile phones. The system architecture to support the indexing and retrieval of images in our data and query model over mobile devices is described. As the popularity of micro-browsers on mobile phones capable of retrieving XTHML-MP pages over WAP gateways increase, it provides an ideal opportunity for us. We can bridge the wireless world with the Internet world seamlessly. A simulation of the concepts using XTHML-MP with WAP 2.0 is provided.

Visualizing Pain Data for Wheelchair Users: A Ubiquitous Approach (pp161-177)
        T. Serif, G. Ghinea and A.O. Frank
We describe a wireless enabled solution for the vizualisation of pain data. Our approach uses pain drawings to record spatial location and type of pain and enables data collection with appropriate time stamping, thus providing a means for the seldom-recorded (but often attested) time-varying nature of pain, with consequential impact on monitoring the effectiveness of patient treatment regimes. Moreover, since the implementation platform of our solution is that of a Personal Digital Assistant (PDA), data collection takes place ubiquitously, providing back pain sufferers with mobility problems (such as wheelchair users) with a convenient means of logging their pain data and of seamlessly uploading it to a hospital server using WiFi technology. Stakeholder results show that, notwithstanding problems related to PDA data input, our approach is generally perceived to be an easy to use and convenient solution to the challenges of anywhere/anytime data collection.

Back to JMM Online Front Page