WWW.DISSERS.RU


...
    !

Pages:     | 1 |   ...   | 18 | 19 || 21 | 22 |   ...   | 54 |

Only a SMS server and a agreement with an SMS provider is needed to implement this service. There are many companies that are actually offering such service. Even a Java implementation could be coded but it is needed a special agreement with the SMS provider in order to deliver all the messages to a given IP. SMS is not a particularly good technology for games, because it is dependent on text entry by the user, and thus is, in essence, a command-line environment. It is also expensive for a game of any depth, since a mere 10 exchanges with the server will cost a user 1 dollar or more. Although the deployment of Multimedia Message Service (MMS) technology makes message-based games more appealing, this is still not a great gameplay environment.

A more complex service than SMS can be implemented if devices support HTTP connections. All new devices are capable of such services. This is an approach similar to WWW services in personal computers. A server is necessary in order to control all the information since the IP address is usually dinamic. Java Servlets technology is an usefull tool to implement such services.

This service can use WML and WMLScript to connect to a server or use J2ME to stablish HTTP connections. In any case schema is similar to normal web pages. A real-time advergame can be achieved with this method since it does not depend on the message delivering just as in the previous section. The information is sent in real-time.

This is a flexible method since only a browser is needed (or a JVM). The main advantage is the graphical user interface. But no P2P communication can be carried out. Either version of WAP offers a friendlier interface than SMS, and is generally less expensive for consumers who pay for airtime only, rather than by the message. But it is a static browsing medium; little or no processing can be done on the phone itself, and all gameplay must be over the network, with all processing performed by a remote server.

Socket use gives J2ME developers the flexibility to develop all kinds of network applications for wireless devices.

However, not every wireless manufacturer supports socket communication in MIDP devices, which means that wireless applications developed using sockets could be limited to certain wireless devices and are less likely to Fourth International Conference I.TECH 2006 be portable across different types of wireless networks. To use a socket, the sender and receiver that are communicating must first establish a connection between their sockets. One will be listening for a request for a connection, and the other will be asking for a connection. Once two sockets have been connected, they may be used for transmitting data in either direction. All today devices are using an embedded JVM, tipically supporting J2ME version 2.0 (with sockets). Main advantages of this implementation are:

Sockets management.

2D and 3D graphical {\it APIs}.

Persinstent storage on the client.

The use of sockets is usefull when dealing P2P services. The only problem is that the IP address is dinamically assigned to the client so a server is needed. This P2P service needs to send some information to an advergame server to keep track information of the players. This solution is the best one since with sockets all previous schemas can be implemented.

Conclusions Advergame is a new marketing concept that brings users a way to interact with others and also to participate in quizs. User information can be update in a database in order to send push messages or does mailing while the client in playing some game. Some advertisements can appear in the game or even play with advertisements.

The user can win prizes to keep his attention.

This paper has presented some tecnologies that can be used to develop an advergame. Java services are the best solution since it is a portable solution and all today devices have an embedded virtual machine.

Bibliography [1] Blockdot (2001). Advergaming 101. Available online: http://www.blockdot.com/advergaming/stats.cfm.

[2] Chen, J., Ringel, M. (2001). Can advergaming be the future of interactive advertising Fast Forward. Available online:

http://www.kpe.com.

[3] March, T. (2001, Spring). How to bag the elusive human attention span. Digitrends. Available online:

http://www.digitrends.net/marketing/13639_16525.html.

[4] Pintak, L. (2001, May 23). Its not only a game: Advergaming set to become a billion dollar industry. Available online:

http://www.turboads.com/richmedia news/2001rmn/rmn20010523.shtml.

[5] Rodgers, A. L. (2002, January). Game theory. Available online: http://www.fastcompany.com/build/build feature/yaya.html.

[6] YaYa (2002a). Why games Available online: http://www.yaya.com/why/index why.html.

[7] YaYa (2002b). YaYa creates viral Internet games that build brands and drive revenue. YaYa online press kit. Available online: http://reports.yaya.com/presskit.pdf.

Authors' Information Eugenio Santos Menendez. Dpto. Organizacin y Estructura de la Informacin. Escuela Universitaria de Informtica de la Universidad Politcnica de Madrid, Ctra. Valencia, km. 7, 28031 Madrid (Spain);

e-mail: esantos@eui.upm.es Rafael Gonzalo Molina Dpto. Inteligencia Artificial. Facultad de Informtica de la Universidad Politcnica de Madrid; Boadilla del Monte, Madrid (Spain); e-mail: rgonzalo@fi.upm.es Francisco Gisbert - Dpto. Lenguajes, Sistemas e Ingeniera del Software. Facultad de Informtica de la Universidad Politcnica de Madrid; Boadilla del Monte, Madrid (Spain); e-mail: fgisbert@fi.upm.es Knowledge Engineering MODELING AND ANNOTATING THE EXPRESSIVE SEMANTICS OF DANCE VIDEOS Balakrishnan Ramadoss, Kannan Rajkumar Abstract: Dance videos are interesting and semantics-intensive. At the same time, they are the complex type of videos compared to all other types such as sports, news and movie videos. In fact, dance video is the one which is less explored by the researchers across the globe. Dance videos exhibit rich semantics such as macro features and micro features and can be classified into several types. Hence, the conceptual modeling of the expressive semantics of the dance videos is very crucial and complex. This paper presents a generic Dance Video Semantics Model (DVSM) in order to represent the semantics of the dance videos at different granularity levels, identified by the components of the accompanying song. This model incorporates both syntactic and semantic features of the videos and introduces a new entity type called, Agent, to specify the micro features of the dance videos. The instantiations of the model are expressed as graphs. The model is implemented as a tool using J2SE and JMF to annotate the macro and micro features of the dance videos. Finally examples and evaluation results are provided to depict the effectiveness of the proposed dance video model.

Keywords: Agents, Dance videos, Macro features, Micro features, Video annotation, Video semantics.

1. Introduction Dance data is essentially multimedia by nature consisting of visual, audio and textual materials. Dance video modeling and mining depends significantly on our ability to recognize the relevant information in each of these data streams. One of the most challenging problems here is the modeling of the dance video semantics such that the relevant semantics are consistent with the perception of the real world.

The classical and folk dances are the real cultural wealth of a nation. In India, the most important classical dances are Bharathanatyam, Kadak, Kadakali, Kuchipudi and Manipuri (Saraswathi, 1994). Traditionally, dance learners perform dance steps by observing the natural language verbal descriptions and by emulating the steps of the choreographers. Therefore, the properly annotated dance videos will help the present and future generations to learn dance themselves and minimize the physical presence of the choreographers.

Notations are used everywhere and are most important for the dancers to communicate the ideas to the learners.

They use graphical symbols such as vertical lines, horizontal lines, dots, triangle, rectangle etc, to denote body parts actions on paper. Labanotation (Hutchinson, 1954) and Banesh (Ann, 1984) have been the frontier notational systems to record the dance movements or dance steps. Many western dances are using Labanotation to describe dance steps. However, many choreographers still follow the traditional way of training their students using natural language descriptions, because of the very few recording experts and inherent complexity of reading and understanding the symbols. Moreover, all Indian dances have unique structure and no common notational structure exists, apart from wire-frame stick diagram representing a dance step. Due to lack of notations, it is evident that the complexity of modeling the dance video semantics is relatively high.

Since the dance steps were archived in paper form and many classical dances lack notations, this kind of archival of dance becomes impossible even today. With the advances in digital technologies (Dorai, 2002) nowadays, magnetic tapes and disks record dance presentations efficiently. But, searching a dance sequence from these collections is not efficient, because of the huge volume of video data. The solution is to build a dance video information system so as to preserve and query the different dance semantics like, dance steps, beyond the spatio-temporal characteristics of the dancers and their dance steps.

The dance video database system requires an efficient video data model to

Abstract

the semantics of the dance videos. To be more precise, the dance video data model should:

Fourth International Conference I.TECH 2006 abstract the different dance video semantics such as dancers, dance steps, agents (i.e., body parts of the dancers), posture, speed of dance steps, mood, music, beat, instrument used, background sceneries and the costume. More importantly, the spatio-temporal characteristics of the dancers must be incorporated in the model;

capture the structure of the dance videos such as shot, scene and compound scene abstracting the different components of the accompanying song.

This paper addresses two related issues: modeling the semantics of the dance videos and annotating the dance steps from the real dance videos. The dance video semantics model represents the different types of dance semantics in a simple, efficient and flexible way. The annotation tool manually annotates the semantics (as macro and micro features) for further query processing and video mining. The main contributions of this paper are as follows:

We propose a generic video data model to describe the dance steps as video events;

We introduce the Actor entity in order to store the event specific roles of a video object. That is, an actor entity describes the context dependent role of the video object;

We introduce the Agent entity to describe the context dependent action that is associated with the actor entity;

We develop a tool that implements the dance video model in order to annotate the different dance semantics.

The rest of the paper is organized as follows: Section 2 presents some related works on video data models.

Section 3 describes the different semantics of the dance videos. The DVSM for the dance video is introduced in Section 4. Section 5 illustrates the implementation of the DVSM using Java technologies. The proposed video model is evaluated against a set of conceptual and semantic quality factors in Section 6. Finally, Section concludes the paper.

2. Related Work Video data modeling is an important component of the dance video database system, as it abstracts the underlying semantics of the dance. This section briefly reviews some of the existing video modeling proposals and discusses the applicability to dance videos.

Colombo(1999) classifies the content-based search as semantic level search (e.g. objects, events and relationships) and low level search (e.g. color, texture and motion). They call the corresponding systems as first and second generation visual information systems. Several key word based techniques are applied to semantic search models, such as OVID (Oomoto, 1993), AVIS (Adali, 1996), Layered model (Koh, 1999) and Schema less semantic model (Al Safadi, 2000). Second generation systems provide automatic tools to extract low level features and subsequently semantic search is performed. Some of these systems include, but not limited to QBIC, Virage, VisualSEEK, VideoQ, VIOLONE, MARS, PhotoBook, ViBE, and PictHunter (Smeulders, 2000;

Antani, 2002). However, these systems are either based on textual annotations or purely low level features, but not incorporating the other one.

In (Shu, 2000), Augmented Transition Network based semantic data model is proposed. The ATN models the video based on scenes, shots and key frames using strings as a sequence of characters. The string representation is used to model the spatial and temporal relationships of each object (moving and static) in a shot of the traffic video. Since the semantic features of dance videos are complex, the entire scene or shot cannot be abstracted in a single string.

Translucent markers, reflector costumes, special sensors and specialized cameras are used to capture and track human body parts movements in some applications such as aerobics, traffic surveillance, sign language, news and sports videos (Vendrig, 2002). In order to record and analyze the dance steps of a dancer, based on this technique requires a special translucent markers or reflector costumes for the dancers. However, dancers do not Knowledge Engineering prefer to use these costumes as these costumes hide the dancers make-ups and costumes. Moreover, these markers and reflectors prohibit the realism, affect dancers comfort as well as reduce the focus or concentration of the dancers. Hence, automatic analysis of dance steps to extract the semantics of the dance steps is very complex.

Recently, the extended DISIMA(Lei Chen, 2003) model expresses events and concepts based on spatio-temporal relationships among salient objects. However, the required dance video database model has to consider not only salient objects, but all objects such as instruments, costumes, background and so on.

Pages:     | 1 |   ...   | 18 | 19 || 21 | 22 |   ...   | 54 |



2011 www.dissers.ru -

, .
, , , , 1-2 .