I completed my Ph.D. at Carnegie Mellon's Robotics Institute. My thesis on "Image and Depth Coherent Surface Description" presents a unified framework to model natural environments using images and depth. The framework reconstructs accurate 3-D models that are coherent with a sequence of observed images and depth data. The data is collected from Unmanned Aerial Vehicles (UAVs) such as autonomous helicopters. I was advised by Prof. Takeo Kanade and Dr. Omead Amidi.

I research computer vision and design algorithms that allow machines to interact with the world using visual cues. I look forward to the day when we build machines that have perceptive, cognitive, and predictive abilities.


EigenFairing: 3D Model Fairing with Image Coherence, P. Mishra, O, Amidi, and T. Kanade, British Machine Vision Conference (BMVC),Kingston, London, 2004. [pdf]

 

 



Key Words: Computer vision, computer graphics, multiple-view geometry, 3-D modeling, image synthesis, visual servoing, autonomous navigation, unmanned aerial vechicles.