What's New:
I am Ph.D. student in the Computer Vision Lab at the University of Maryland under supervision of Yiannis Aloimonos.
Office: AVW 4470 :: Tel.: 301 405 1762 :: jneumann at cfar.umd.edu
Research Statement:
My main research interests deal with the questions how to represent the visual information captured by many imaging devices simultaneously and how to represent and extract the spatio-temporal structure of the world that is observed.
Plenoptic Video Geometry.
Nowadays, cameras and computing resources become cheaper and cheaper which enables us to capture video sequences using hundreds of cameras at the same time. These camera arrangements should not be modeled as sets of discrete cameras anymore, but one should study directly the continuous space of all light rays as described by the plenoptic function and interpret this multitude of video sequences as spatio-temporal samples in plenoptic space. By extending conventional signal processing techniques to the space of time-varying light rays, I study the Plenoptic Video Geometry, that is the geometric and differential structure of all the visual information that an imaging sensor can possibly capture.
Polydioptric Camera Design.
One result I found is that for a polydioptric
camera, that is a generalized camera that captures a multi-perspective subset of the space of light rays, the 3D motion estimation problem becomes scene independent.
I used this result and my analysis of the Plenoptic Video Geometry
to devise a classification of cameras
with regard to the structure from motion problem which I use as
a foundation to design new imaging geometries that are optimal for
the task of 3D motion estimation.
Spatio-temporal Reconstruction from Video.
Along another line of work, I am also studying how to represent and extract the spatio-temporal structure of complex non-rigidly deforming objects from many (discretely or continuously spaced) viewpoints. Some initial results can be found here, where I reconstructed a talking head using spatio-temporal stereo information and multi-resolution subdivision surfaces. For more detailed information about my research please take a look at my research page.