http://www.cse.ohio-state.edu/lair/knowsys.html

The Knowledge Systems group at the LAIR is concerned with making computers smart by giving them knowledge about the world and methods of using knowledge to solve problems. Thus knowledge representation and problem solving are two key themes in the group's work. The strategy adoped by the Knowledge Systems group is to focus on complex real-world tasks. Cognitive architectures, and abstract principles of knowledge-based reasoning and problem solving are considered and utilized.

http://www.cse.ohio-state.edu/cvl

Our research examines the perceptual recognition of human motion from computational and cognitive viewpoints. In particular, how do we bestow into computers the ability to recognize our movements in the world that have meaning, intention, and expression? We approach this problem from a computational standpoint with the belief that there exist structures, regularities, and modes within human movements that offer a reliable means to constructing representations for robust motion categorization. The applied significance for this work most closely relates to automatic visual surveillance and monitoring, video content understanding, perceptual user interfaces, and human-computer interaction.

http://www.cse.ohio-state.edu/pnl

This lab conducts research on understanding neurocomputational mechanisms underlying perceptual processes and on building effective algorithms for solving real-world problems related to machine perception. We view these two aspects of our goal as intimately related, as we believe that information-processing mechanisms of the brain as the product of millions of years of evolution represent the optimum or near-optimum computational algorithms, and conversely the computational algorithms that ultimately work for modeling perception are close to what actually are used by the brain.

http://www.cse.ohio-state.edu/slate

Our research goal is to build software that can do a better job at understanding human languages. Natural language understanding technology can be applied in two ways. First, as an interface to computer applications, for example, a spoken interface to a scheduling system. Second, as the data used by an application, such as in translating documents from one language to another. We currently focus on the problem of interpreting referring expressions (such as noun phrases and pronouns) because this is a central problem for language understanding that currently has no satisfactory solution.