Deformable Models for Computer Vision
Project Award Date: 0000-00-00
The principal objective of this research project is the development and application of deformable models for computer vision. In particular: 1) to improve computational methods for fitting deformable models to features of interest in 2D and 3D images, 2) to investigate how deformable models can dynamically adapt surface characteristics to best fit the underlying structures in images, 3) to evaluate the effectiveness of deformable models for segmentation and object recognition tasks. Current application areas for our deformable models include:
- fitting cell boundaries in 3D fluorescent microscopy data,
- fitting body and lung outlines in 3D nuclear medicine images,
- tracking non-rigid motion of objects in biological image sequences,
- surface reconstruction in a multisensor data fusion application.
Primary Sponsor(s): National Science Foundation, Robotics and Machine Intelligence