Dr. Melanie A Sutton

The University of West Florida (UWF) is a relatively young institution, founded in 1967 as an upper-level institution enrolling juniors, seniors, and graduate students. The university became a full 4-year undergraduate institution and admitted its first class of freshmen in 1983. A significant number of undergraduate students transfer to UWF after beginning their studies elsewhere; nearly 50% of students who graduate with a bachelor's degree from UWF matriculate with an A.A. The university admitted the first cohort of graduate students in an Ed.D. program in 1996. The university currently enrolls approximately 11,200 students. This enrollment allows for relatively small class sizes (the average class enrollment is 25 students). The student population is drawn largely from Florida (89% of students are Florida residents, and 11 % are non-Florida residents), although the student population includes students from all 50 states and 92 countries. Approximately 23% of UWF students identify themselves as members of a cultural/ethnic minority group.

This paper presents the framework of the new context-based reasoning components of the GRUFF (Generic Recognition Using Form and Function) system. This is a generic object recognition system which reasons about and generates plans for understanding 3-D scenes of
objects. A range image is generated from a stereo image pair and is provided as input to a multi-stage recognition system. A 3-D model of the scene, extracted from the range image, is processed to identify evidence of potential functionality directed by contextual cues. This recognition process considers the shape-suggested functionality by applying concepts of physics and causation to label an object's potential functionality. The methodology for context-based reasoning relies on determining the significance of the accumulated functional evidence derived from the scene. For example, functional evidence for a chair or multiple chairs along with a table, in set configurations, is used to infer the existence of scene concepts such as "office" or "meeting room space." Results of this work are presented for scene understanding derived from both simulated and real sensors positioned in typical office and meeting room environments.

This work summarizes research which uses various aspects of functionality as a means to achieve generic object recognition. We examine several approaches which attempt to integrate computer vision and robotics for the purpose of achieving recognition of functional classes of objects. Such integrated approaches include those which incorporate both the knowledge of the potential functionality of an object and the steps to confirm said functionality through interaction. An overview of our system, GRUFF-I (Generic Recognition Using Form, Function and Interaction), is presented. This system reasons about and generates plans for interaction with 3-D shapes from the categories furniture and dishes.