Object Recognition and Segmentation

Abstract

Both the ability to organize signal samples into symbolic primitives and the ability to recognize groups of symbolic primitives are necessary for machine perception. Just as a person must identify phonemes to form words to understand a sentence, a computer vision system must identify 3-D surfaces belonging to 3-D objects to understand a 3-D scene. That is, surface segmentation and object recognition are fundamental tasks in image understanding. Although the goal is to devise a data-driven algorithm that extracts surface primitives from images without knowledge of higher level objects, a systems approach is taken and the entire problem is examined first before attempting to solve a part of it.