The capabilities for representing and reasoning about three-dimensional objects are essential for . knowledge -based, 30 photo-interpretation systems that combine domain knowledge with image processing, as demonstrated by such systems as 30 Mosaic and Acronym. Three-dimensional representation of objects is necessary for many additional applications such as robot navigation and 30 change detection. Geometric reasoning is especially important, since geometric relationships between object parts are a rich source of domain knowledge. A practical framework for geometric representation and reasoning must incorporate projections between a 2D image and a 3D scene, shape and surface properties of objects, and geometric and topological relationships between objects. In addition, it should allow easy modification and extension of the system's domain knowledge and be flexible enough to organize its reasoning efficiently to take advantage of the current available knowledge. We are developing such a framework, called the 3D FORM (Frame-based Object Recognition and Modelling) System. This system uses frames to represent objects such as buildings and walls, geometric features such as lines and planes, and geometric relationships such as parallel lines. Active procedures attached to the frames dynamically compute values as needed. Since the order of processing is controlled largely by accessing objects' slots, the system performs both top-down and bottom-up reasoning, depending on the current available knowledge. The FORM system is being implemented using the CMU-buitl Framekit tool in Common Lisp [3]. Examples of interpretation using a simple model of building as rectangular prism are presented. l This research was sponsored by the Defense Advanced Research Projects Agency (DOD), ARPA Order No. 4976, monitored by the Air Force Avionics Laboratory under Contract F33615-84-K-1520. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Defense Advanced Research Projects Agency or of the U. S. Government.
[1]
S. Shafer.
Shadow geometry and occluding contours of generalized cylinders (artificial intelligence)
,
1983
.
[2]
Takeo Kanade,et al.
Incremental Acquisition of a Three-Dimensional Scene Model from Images
,
1982,
IEEE Transactions on Pattern Analysis and Machine Intelligence.
[3]
Rodney A. Brooks,et al.
Symbolic Reasoning Among 3-D Models and 2-D Images
,
1981,
Artif. Intell..
[4]
Nils J. Nilsson,et al.
Artificial Intelligence
,
1974,
IFIP Congress.
[5]
Paliath Narendran,et al.
Reasoning about three dimensional space
,
1985,
Proceedings. 1985 IEEE International Conference on Robotics and Automation.
[6]
Takeo Kanade,et al.
Incremental Reconstruction of 3D Scenes from Multiple, Complex Images
,
1986,
Artif. Intell..
[7]
Shang-Shouq Vincent Hwang,et al.
Evidence accumulation for spatial reasoning in aerial image understanding (expert system, computer vision)
,
1983
.
[8]
Joseph L. Mundy,et al.
A Multi-Level Geometric Reasoning System for Vision
,
1988,
Artif. Intell..
[9]
Alan K. Mackworth.
Interpreting Pictures of Polyhedral Scenes
,
1973,
IJCAI.