Ivica Arsov

Optimized MPEG-4 animation encoder for motion capture data

This paper presents first compression results on using the MPEG-4 BBA standard for encoding motion capture data. We first introduce a detailed description of the main compression mechanisms used in recently BBA standard: prediction, frequency transform, quantization and entropy encoder and discuss the theoretical range of encoding performances. Then we introduce an optimized BBA encoder including also a key-frame reduction mechanism and show the compression results on animation files from typical motion capture data-base. We compare the results with the ones reported in literature showing the advantages of the MPEG-4 BBA in terms of bit-rate, complexity and range control.

A server-assisted approach for mobile-phone games

While the processing power of mobile devices is continuously increasing, the network bandwidth is getting larger and the latency is going down, the heterogeneity of terminals with respect to operating systems, hardware and software makes it impossible to massively deploy mobile games. We propose a novel architecture consisting of (1) maintaining on the terminal only standardized operations (2) using a remote server for the games logic and (3) updating the local scene by using compact commands. By identifying the game tasks that should remain local, we first analyze the performances of MPEG-4 3D graphics standard. We implemented an MPEG-4 player able to decode and visualize 3D and extended it for handling game content. We propose a client-server architecture and a communication protocol that ensure similar user experience while using a standard player - expected to be available on mobile devices in the near future.

Adaptable client-server architecture for mobile multiplayer games

Most of the multiplayer games available online are based on a client-server architecture because this architecture gives better administration control to the game providers. Besides controlling the account and payment information of the players, this architecture also prevents players from cheating as all the game logic is executing on the centralized server. We proposed a server assisted approach for mobile games in [2]. However, because of the varying and high latency of wireless networks and of the changing consistency requirements during the game play, it is difficult to keep the user experience highly interactive in client-server architecture. This paper presents an adaptive hybrid client-server architecture which changes its behavior according to network and game environment variations to improve game state consistency across different mobile terminals. The server applies consistency mechanism on its side, as in the traditional client-server architecture and dynamically switches to apply a client side consistency mechanism when inconsistencies occur at the client side because of the change in network conditions and/or game requirements. We have evaluated our approach on a car racing game. The results show that we can obtain an improved global consistency under a high and varying latency network using our dynamically adaptable approach.

Mobile mixed reality games creator based on MPEG-4 BIFS

In this paper, we propose an authoring tool for creating mobile mixed reality games (MMRGs), referred to as M2RGC (Mobile Mixed Reality Game Creator). Goal is to transform the game creation process in a task that is as interesting as playing the game itself. Therefore we offer a user-friendly interface through which different MMRGs can be created by defining the objects used in the game, their behavior and interaction with sensors (e.g. GPS, compass). The tool itself uses the power of MPEG-4 BIFS[1] capabilities, such as the possibility of grouping multimedia assets in a scene. We build a wrapper over these MPEG-4 BIFS functionalities allowing anyone to discover the experience of MMRGs creation. Concerning the client runtime for playing the game on mobile phones, we deploy a MPEG-4 player on the Windows Mobile and Android platforms.

ONLINE ANIMATION SYSTEM FOR PRACTICING CUED SPEECH

This paper presents a set of information and communication technologies developed with the goal to improve the practice of cued speech. They are based on three-dimensional (3D) graphics and are covering the entire end-to-end content chain: production, transmission, and visualization. Starting from defining a set of potential applications, the requirements of the targeted system for cued speech are set. The research and development path takes into account high-quality animation, real-time constraints, personalization, user acceptability and, equally important, the easiness and feasibility of the deployment. The latter led to a strong orientation toward open standards. The original components of the system include 3D graphics and animation encoders, streaming servers, and visualization engines. The core technology is validated in two real-time applications: a web service for text to animation conversion and a chat service supporting two or more users.