Generalized Image Acquisition and Analysis
Interactive Geometry-Aware Segmentation for the Decomposition of Kaleidoscopic Images
Mirror systems have recently emerged as an alternative low-cost multi-view imaging solution. The use of these
systems critically depends on the ability to compute the background of a multiply mirrored object. The images
taken in such systems show a fractured, patterned view, making edge-guided segmentation difficult. Further,
global illumination and light attenuation due to the mirrors make standard segmentation techniques fail.
We therefore propose a system that allows a user to do the segmentation manually. We provide convenient tools
that enable an interactive segmentation of kaleidoscopic images containing three-dimensional objects. Hereby, we
explore suitable interaction and visualization schemes to guide the user. To achieve interactivity, we employ the
GPU in all stages of the application, such as 2D/3D rendering as well as segmentation.
Projects
Discovering the Structure of a Planar Mirror System from Multiple Observations of a Single Point
CVPR 2013
Abstract
We investigate the problem of identifying the position of a
viewer inside a room of planar mirrors with unknown geometry
in conjunction with the room’s shape parameters. We consider
the observations to consist of angularly resolved depth
measurements of a single scene point that is being observed
via many multi-bounce interactions with the specular room
geometry.
Applications of this problem statement include areas such as
calibration, acoustic echo cancelation and time-of-flight
imaging. We theoretically analyze the problem and derive
sufficient conditions for a combination of convex room
geometry, observer, and scene point to be
reconstructable. The resulting constructive algorithm is
exponential in nature and, therefore, not directly
applicable to practical scenarios.
To counter the situation, we propose theoretically devised
geo- metric constraints that enable an efficient pruning of
the solution space and develop a heuristic randomized search
algorithm that uses these constraints to obtain an effective
solution. We demon- strate the effectiveness of our
algorithm on extensive simulations as well as in a
challenging real-world calibration scenario.
Bibtex
@inproceedings{Reshetouski:13,
author = {Ilya Rehsetouski and Alkhazur Manakov and Ayush Bhandari and Ramesh Raskar and Hans-Peter Seidel and Ivo Ihrke},
title = {Discovering the Structure of a Planar Mirror System from Multiple Observations of a Single Point},
booktitle = {Proceedings of CVPR},
year = 2013,
pages = {xx--yy},
}
author = {Ilya Rehsetouski and Alkhazur Manakov and Ayush Bhandari and Ramesh Raskar and Hans-Peter Seidel and Ivo Ihrke},
title = {Discovering the Structure of a Planar Mirror System from Multiple Observations of a Single Point},
booktitle = {Proceedings of CVPR},
year = 2013,
pages = {xx--yy},
}