People

My research interests are in computer vision and computer graphics. Primarily, I am interested in developing physics-based models and algorithms for scene interpretation in vision and real-time rendering of physical effects in graphics.

I am interested in using computer vision to solve real world problems, both for human assistive devices, as well as autonomous robotic applications. Specifically, I am currently focused on scene understanding and systems that assist drivers.

My research interests are in Physics-based Computer Vision and Graphics. In particular, I am interested in understanding how light interacts with participating media (ocean water, smoke, dust etc.) and the resulting effects on the image formation process. Most of the current robot-vision algorithms assume 'perfect-weather' conditions. Hence, they tend to break down in presence of imperfect weather conditions like fog, mist etc. By modeling the interaction of light with such participating media, I want to be able to make vision algorithms robust even to bad weather and underwater scenarios.

My research interests focus on computer vision and graphics based medical applications, especially on computer assisted minimally invasive surgury. I am advised by Branislav Jaramaz and Srinivasa Narasimhan and working with the team in ICAOS.

My research interests span all combinations of lights, cameras and computers for application in vision and graphics. Specifically my work has three areas of focus. First, modeling a scene's visual appearance to find physics-based invariants for data-driven and learning algorithms. Second, applying active illumination to recover geometry and create novel visual experiences. Third, exploiting vision algorithms and psychophysics techniques to create, enhance and edit stereographic content.

My research interests are in the area of the 3D scene modeling, image processing and computer graphics. My current project is on image-based scene modeling using volume representation. The current target of modeling is individual human hair for the purpose of photorealistic image synthesis, although we can generalize the method to a broader scope.