Research
/
Research Areas
Vision and Graphics

We design, model, and build systems that combine sensors, displays, and novel optical elements to enable new functionality in cameras and displays for applications in medical, astronomical, and scientific imaging.

The same physical models we use to render images (e.g. camera, material, geometry, and lighting models) can also be used to develop algorithms that mine today's vast collections of digital images for richer scene understanding, paving the way towards smarter, more intelligent cameras that augment our visual perception.

We use analytical tools and machine learning models to train computers to interpret the visual world through tools such as feature analysis, image segmentation, object recognition, edge detection, pattern detection, image classification, and feature matching.

Computer vision and graphics have a natural synergy with many other fields in computer science including robotics, human-computer interaction, and machine learning. As a result, many of the algorithms we develop have broad applications that extend beyond simulation, optics, image processing, modeling, and visualization.