Please join me in congratulating , Luddy Professor of Computer Science in the Luddy School of Informatics, Computing and Engineering, on receiving a patent from the U.S. Patent and Trademark Office for the development of a depth estimation method to reconstruct 3D images.
Professor Crandall’s patent, awarded in May of this year, was developed with Md Alimoor Reza, a former postdoctoral associate at Luddy. The patent focuses on computer vision—specifically, how to reconstruct 3D models from 2D images—and has broad applicability in numerous fields including medical imaging and diagnostics, sports performance analysis, construction and architecture, autonomous driving, and robotics.
A of the patent award notes, “Current methods of reconstructing 3D images require identifying a large amount of image data, while Crandall’s and Reza’s depth estimation method uses machine learning to reconstruct a 3D image from a single or small amount of 2D images. The method also enhances the accuracy of the 3D scene reconstruction by diagnosing errors.”
Congratulations, Professor Crandall!
