CS Colloquium Series

Image super-resolution (SR) creates an enhanced high-resoultion (HR) image using multiple low-resolution (LR) images of the same object. A typical image formation model introduces blurring, aliasing, and added noise. Super-resolution is designed to jointly reduce or remove all three.

For both humans and machines, the ability to learn and recognize the semantically meaningful contents of the visual world is an essential and important functionality.

Second Life is a massively-multiplayer 3D virtual world, unique in that it is completely created by its residents. Via a mix of technology, ownership, markets, and community, Second Life has dispelled many myths about the quantity and quality of user-created content.

Testing with manually generated test cases is the primary technique used in industry to improve reliability of software--in fact, such testing is reported to account for over half of the typical cost of software development.

Computer security research is a broad field, with research efforts ranging from the design and analysis of low-level cryptographic building blocks to the design and analysis of complex and socially important systems.

Computers continue to get broken into, so intrusion analysis is a part of most system administrators'job description. System administrators must answer two main questions when analyzing intrusions: "how did the attacker gain access to my system?", and "what did the attacker do after they broke in".

I will introduce several low-level vision tasks, such as image reconstruction and optical flow estimation, and show how they can be approached in a unified way as Bayesian inference problems. One key component of these Bayesian approaches is modeling the prior distribution.

We have rather detailed, if tentative, information about how organisms learn from experience to choose better actions. But it is much less clear how they arrange to obtain this experience.

Local image features have emerged as a powerful way to describe images of objects and scenes. Their stability under variable image conditions is critical for success in a wide range of recognition and retrieval applications.

Automatically extracting structured facts from unstructured text is a key step toward natural language understanding. Many researchers study this problem, typically in the context of text collections such as newsfeeds or the web.