Representation, Modeling, and Recognition of Outdoor Scenes

摘要

The goal of this project is to advance the state-of-the-art in scene interpretation for autonomous systems that operate in natural terrain. In particular, techniques are being developed for representing knowledge about complex cultural and natural environments so that a computer vision system can successfully plan, navigate, recognize, and manipulate objects, and answer questions or make decisions relevant to this knowledge. The results to date include the development of new representations and techniques for rapidly modeling terrain from multiple images, and for the recognition and reliable labeling of such scene attributes and components as color, texture, shadows, and a variety of linear structures (skyline, ridgelines, road, etc.). The most recent results are detailed in three papers included as appendices to this report.