4D model construction and updating
Choice of Geometry: The aim is to determine the optimal camera (both intensity
and range) geometry for capturing time-varying or stationary scene structures,
on a moving platform.
Choice of Imaging Technology: The aim is to determine the best strategy
in terms of the number and type of cameras used with the objective of generating
photo-realistic 4D scene models.
Choice of Representation: Find the optimal trade-off between compactness
to meet the bandwidth requirements for a mobile visualization system and
the fidelity to photo-realistic representations.
Choice of Models: determine the trade-off between various models such as voxel-based
or surface-based models in terms of their performance in rendering and
compression.
Compression: Extend 2D video compression techniques to 3D video compression.
Geo-Registration and Tracking
Geo-registration of remotely-sensed data both with and without base
map.
Image to image and image to site-model registration based on GPS, inertial
navigation systems (INS) and camera calibration.
Investigate methods for jointly optimizing the correlation search for height
and texture variations.
Robust image tracking using Kalman filtering and robust statistics.
Visualization System/Architecture
Span the whole range of visualization systems from command posts and
multiple screen displays and immersive environments, to workstations and
portable systems.
Dynamic, universal data structures with fast updates for real-time global
terrain and 3D feature visualization, with interactive access to accurate
views and blending of multiple images.
High resolution 3D data viewing capability, with a continuous level of
detail.
Intelligent data mining for visualization and monitoring.
Uncertainty Evaluation & Visualization
Developing methodologies for evaluating uncertainty due to imperfect,
imprecise and incomplete data.
Developing methodologies for evaluating uncertainty in 4D model building
and geo-registration.
Formal and systematic study of uncertainty in mobile information systems
and data fusion systems.
Suitable and effective presentation of uncertainty for time-critical decision-making applications.
Thorough study and implementation of uncertainty visualization and management
in a distributed network in terms of its computation, fusion of information,
visual representation and validation.