EPSRC logo

Details of Grant 

EPSRC Reference: GR/H83676/01
Principal Investigator: Taylor, Professor CJ
Other Investigators:
Researcher Co-Investigators:
Project Partners:
Department: Imaging Science & Biomedical Eng
Organisation: Victoria University of Manchester, The
Scheme: Standard Research (Pre-FEC)
Starts: 01 February 1993 Ends: 31 July 1996 Value (£): 300,981
EPSRC Research Topic Classifications:
Image & Vision Computing
EPSRC Industrial Sector Classifications:
Related Grants:
Panel History:  
Summary on Grant Application Form
Create composite shape and grey-level models for objects of variable appearance Develop tools for acquiring object appearance models from example images Develop efficient methods for locating matches between models and image data Extend the modelling and matching methods to deal with 2D images of 3D objects Test the ability to model, locate and recognise objects in a range of application domainsProgress:The project is intended to build on earlier SERC-funded work on the use of statistical shape models in automated image interpretation. The aim is to use prior knowledge of the objects and structures which are expected to appear in images to arrive automatically at an interpretation in which all the relevant image components have been located and labelled. The use of prior knowledge should allow reliable interpretation even if the image evidence is poor due to intrinsically low contrast, clutter, noise or occlusion. We have made substantial progress towards achieving all of the specific project objectives. Our original approach to modelling shape variation, which was linear, has been extended to deal with non-linear constraints on shape and spatial relationships; both a polynomial approximation and a more general neural net approach have been developed. Two different methods of combining grey-level information with shape models have been developed and tested in a range of applications. A set of software tools for creating and testing models have been developed and are already being used by industrial and academic collaborators. Our original scheme for matching shape models to image data has been extended to make use of grey-level models. The efficiency, robustness and accuracy of matching have also been improved considerably by modifying the matching algorithm, particularly by adopting a multi-scale approach. We have taken results from recent work on the recovery of structure from uncalibrated images to extend the modelling approach to deal with 3D objects seen in 2D images from arbitrary viewpoints. This work is at an early stage but opens up the prospect of developing a completely general approach to image modelling and interpretation. Systematic testing of the new methods has been undertaken using images from applications as diverse as face recognition, industrial inspection, traffic monitoring and medical image interpretation. The results demonstrate that the scope of the statistical model-based approach has been extended significantly by our work and that, for applications which were already feasible, significant improvements in performance can be achieved.
Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:  
Further Information:  
Organisation Website: