Next: Scenarios & Dreams
Up: (Intelligent) Agents and
Previous: (Intelligent) Agents and
As a general assumption, in we follow a bottom-up approach to AI,
grounding representation and reasoning models on the psycho-physical and
perceptual aspects. Examples of bottom-up approaches starting from the
signal, perceptual level of representation of musical signals can be found
in [80,8,183].
According to our bottom-up approach to AI and multimedia, we are mainly
interested in the study and development of ``autonomous'' multimedia
systems, that is, autonomous agents characterized by multimodal
interaction with user(s) in an unstructured and evolving
environment [201,320]. In particular, examples of real world
scenarios that are considered in are the following:
- a theatrical machine: a system delegated to manage and integrate
sound, music, and either three-dimensional computer animation of humanoid
figures (e.g., dance movements) or the movement of real autonomous
agents in a theater stage (e.g., a real vehicle on wheels, equipped
with on-board sensors, a computer for the low-level processing of
sensorial data, etc.). In both cases we have an agent which should be
able to move, navigate, react to events happening on stage (e.g.,
actions performed by the actors), to acquire sounds from the environment,
and possibly execute musical tasks;
- a museal machine, based on an autonomous robot, very similar in its
architecture to the theatrical machine, operates in real time in a museum
exhibition area, and is able to welcome, entertain, guide, and instruct
visitors. See [319] for a different AI-based model developed for a
similar museal domain. In both cases, a musical competence is part of the
requirements for the systems.
- interactive systems for the acquisition of human movement (at various
levels of abstraction) and their use in various applications: e.g., use
of this information to control computer tasks going from advanced user
interfaces to entertainment tools.
The main keypoints considered in the design of experimental system
architectures in these application scenarios are: (i) high-level
integration of different modalities and skills; (ii) high level,
multimodal interaction with the real world in real time: for example, such
a system builds up representations and reasons on a realistic model of what
is happening on stage, e.g., for deciding how to interpret or generate a
music object, far more than a simple ``triggering'' mechanism: interaction
therefore means a deeper mechanism than a simple temporal synchronization
of chunks of music or graphics/animation data. Given the complexity of the
problem domain, we deem that a single knowledge representation and
reasoning formalism is not sufficient for all the aspects of such complex
domains.
\
Next: Scenarios & Dreams
Up: (Intelligent) Agents and
Previous: (Intelligent) Agents and
Esprit Project 8579/MIAMI (Schomaker et al., '95)
Thu May 18 16:00:17 MET DST 1995