VEIL: Research in Knowledge Representation for Computer Vision.
Abstract
The VEIL (Vision Environment Integrating Loom) project integrated advanced Knowledge Representation (KR) technology (Loom) with image understanding technology. The major innovations are as follows: 1) Use of declarative knowledge (as opposed to procedural knowledge) in vision systems made extending the recognition capabilities of the software easier. 2) Facilitated integrating high-level vision routines (recognizing sequences of scenes) with low-level routines that recognize picture elements. 3) Declarative knowledge enabled interaction with the system at a level of abstraction appropriate to the domain task. This included associating collateral information with the objects recognized by low-level image understanding programs. Loom is a powerful tool incorporating strong, frame-based representation capabilities, explicit term subsumption, and several powerful reasoning paradigms (including logical deduction, object-oriented methods, and production rules). Loom also provides knowledge representation integrity through consistency checking and truth maintenance. Infusing these facilities into the vision problem area, where strong KR capabilities have not yet been developed, significantly altered and improved the methodology for constructing vision systems. Finally, Loom was interfaced to a variety of vision processing elements to provide a new tool of extended capabilities. The result is a powerful software environment for the development of vision systems.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1998
- Accession Number
- ADA341150
Entities
People
- Behnam Salemi
- Keith Price
- Ramakant Nevatia
- Robert B. Macgregor
- Thomas A. Russ
Organizations
- University of Southern California