|
Image and Data Processing Group
|
|
General Information
Modern society increasingly relies on the collection, processing, and simulation of massive data volumes. While we have been extremely successful in developing these techniques, the current trend towards more realistic physics-based simulations and inferences reveal shortcomings in our ability to handle high-dimensional data volumes. Examples of areas that are struggling with this data deluge include computer graphics and gaming, with consumers demanding more and more realism; as well as seismic and medical imaging, where there is incessant push towards higher resolution images and better inferences on what these images contain. The Image and Data Processing Group aims to develop new techniques to address this challenge.
| |
At UBC, we have a very strong and diverse group of faculty working on various aspects of image and data processing. We have faculty in the fields PDE- and differential-geometry-based image generation and registration, and we also have faculty active in the fields of computational and applied harmonic analysis (wavelets); compressive sensing (a new paradigm in sampling); convex optimisation (one-norm solvers); machine learning; and PDE-constrained optimisation (seismic imaging).
Aside from efforts by individual faculty in these areas, the image and data processing group also includes the industry-supported interdisciplinary research program DNOISE II, which involves faculty from Mathematics, Computer Science, and Earth & Ocean Sciences, and a joined seminar series (see our blog) on stochastic optimisation and compressive sensing. The latter seminar is attended by faculty and students from several departments.
|
|
|
|
TOP
Core Faculty
The Image and Data Processing Group is composed of the following core IAM faculty. Prospective students interested in a research project in Image and Data Processing in the IAM are encouraged to contact one or more of the core faculty as potential supervisors and let them know of their interests.
| |
| Uri Ascher |
Uri is a Professor of Computer Science and a former Director of the IAM (1993-98). The focus of his work is the investigation and promotion of novel, efficient and reliable methods in scientific computation, particularly for approximation problems involving differential equations with constraints. Examples of Uri's specific research areas are robotics, virtual reality, data inversion in geophysics, multibody systems simulation, 3D electromagnetic modelling, image reconstruction, and 3D mesh denoising.
|
| Robert Bridson |
Robert is a Professor of Computer Science and a member of the Imager lab. His work revolves around numerical and geometric algorithms for solving problems in fluid and solid mechanics, including iterative methods for linear systems, discretisation of PDEs in space and time, mesh generation, and collision and contact handling. He often works closely with film studios in applying these algorithms to problems in physics-based animation.
|
| Michael Friedlander |
Michael is a Professor of Computer Science. His research is primarily in developing numerical methods for large-scale optimisation. He is especially interested in issues of convergence analysis, robust software implementation, and applications in signal processing and image reconstruction. His recent contributions include software packages for large-scale sparse optimisation ( SPGL1),
sparse signal reconstruction ( Sparco),
and a linear operator toolbox ( SPOT).
|
| Eldad Haber |
Eldad is a Professor of Mathematics and of Earth and Ocean Sciences. His main field of interest is the development of computational methods for inverse problems with applications to geophysical and medical imaging. The field is interdisciplinary by nature and includes numerical discretisation of partial differential equations, numerical optimisation and robust statistics. Eldad is an NSERC Industrial Research Chair in Computational Geoscience.
|
| Felix Herrmann |
Felix is a Professor of Earth and Ocean Sciences and the director of the Seismic Laboratory for Imaging and Modeling ( SLIM), which is receiving support from oil and gas industry through the SINBAD consortium. Felix is also the principal investigator of the DNOISE project, which is supported by NSERC. Felix' research programme focusses on various aspects of exploration seismology, including seismic-data acquisition, processing, imaging, and inversion.
|
| Özgür Yilmaz |
Özgür is a Professor in the Department of Mathematics. His main research areas are quantisation of redundant expansions, blind source separation, and sparse approximations. He is a co-author of Sparco, a toolbox for testing sparse reconstruction toolbox algorithms.
|
| Alla Sheffer |
Alla is a Professor of Computer Science and a member of the Imager lab. She conducts research in digital shape modeling and geometry processing, with applications to computer graphics and computer-aided engineering. Her work utilises tools from computational and differential geometry, discrete mathematics, and graph theory to generate, manipulate, and edit discrete geometric models.
|
|
|
|
|
TOP
Recommended Courses
Students interested in Image and Data Processing research in the IAM are advised to take the following preliminary, specific and optional courses:
| |
|
|
Preliminary and Foundational Courses
CPSC 302: Numerical Computation for Algebraic Problems
CPSC 303: Numerical Approximation and Discretization
CPSC 314: Computer Graphics
MATH 340: Introduction to Linear Programming
MATH 405: Numerical Methods for Differential Equations
CPSC 542G: Introduction to Numerical Methods
Specific Courses
EOSC 513: Imaging and Estimation with Wavelets
CPSC 524: Computer Graphics: Modeling
CPSC 526: Computer Animation
CPSC 530P: Sensorimotor Computation
MATH 555: Compressed Sensing
Further Options
CPSC 533D: Animation Physics
EOSC 550: Linear Inverse Theory
EOSC 555: Nonlinear Inverse Theory
CPSC 564: Data Mining
|
|
TOP
Sample Publications
Listed below are some of the recent theses and journal publications by IAM students and IAM faculty in the field of Image and Data Processing. Click on the item of interest to open a thesis pdf file, an article on the author's web page, or an article abstract on the journal site.
| |
Eldad Haber, Matthias Chung and Felix Herrmann. An effective method for parameter estimation with PDE constraints with multiple right hand sides. Preprint (2010).
Tyson Brochu, Christopher Batty and Robert Bridson. Matching Fluid Simulation Elements to Surface Geometry and Topology. Proceedings of the Association for Computing Machinery's Special Interest Group on Computer Graphics and Interactive Techniques (2010).
Derek Bradley, Wolfgang Heidrich, Tiberiu Popa and Alla Sheffer. High Resolution Passive Facial Performance Capture. ACM Transactions on Graphics, 29(3) (2010).
Ravish Mehra, Pushkar Tripathi, Alla Sheffer and Niloy Mitra. Visibility of Noisy Point Cloud Data. Computers and Graphics, 34(3), pp. 219-230 (2010).
Rayan Saab and Özgür Yilmaz. Sparse Recovery by Non-Convex Optimization – Instance Optimality. Applied and Computational Harmonic Analysis, 29(1), pp. 30-48 (2010).
Rayan Saab. Compressed Sensing: Decoding and Quantization. Ph.D. Thesis, University of British Columbia (2010).
Tyson Brochu and Robert Bridson. Robust Topological Operations for Dynamic Explicit Surfaces. Accepted to SIAM Journal on Scientific Computing.
Felix Herrmann, Yogi A. Erlangga and Tim T. Y. Lin. Compressive Simultaneous Full-Waveform Simulation. Geophysics, 74, p. A35 (2009).
Gilles Hennenfent and Felix Herrmann. Simply Denoise: Wavefield Reconstruction via Jittered Undersampling. Geophysics, 73(3), p. V19 (2008).
Hagit Schechter and Robert Bridson. Evolving Sub-Grid Turbulence for Smoke Animation. ACM SIGGRAPH Symposium on Computer Animation (2008).
Kees van den Doel, Uri Ascher and Dinesh Pai. Computed Myography: Three Dimensional Reconstruction of Motor Functions from Surface EMG Data. Inverse Problems 24, 065010 (2008).
Hui Huang and Uri Ascher. Surface Mesh Smoothing, Regularization and Feature Detection. SIAM Journal on Scientific Computing, 31(1), pp. 74-93 (2008).
Hui Huang. Efficient Reconstruction of 2D Images and 3D Surfaces. Ph.D. Thesis, University of British Columbia (2008).
|
|
|
|
TOP
|
