Linh Khanh Ha
Research Assistant
Visualization and
Geometric Computing (VGC) group
Scientific Computing and Imaging
Institute
Computer Science Department
University of Utah
Office :
Howell Student Lab
Scientific Computing and Imaging Institute
72
S Central Campus Drive, Room 3750
Salt Lake City, UT 84112
Phone :
(801) 326 - 9208
E-mail: lha@sci.utah.edu
| About | |
|
I'm a Ph.D in Computing Track at the School of Computing, Computer Science Department, University of Utah. I sucessfully defended my thesis on February 25th, 2011. My thesis is about how to build a high performance image processing framework to harness GPU parallel processing power for computational research challenges. My advisor is Claudio T Silva. My co-advisors are Sarang Joshi and Jens Kruger. My research focuses on High Performance Computing, Visualization, Parallel Processing and GPGPU programming. |
|
| Background | |
|
I have a strong math background. I had received several awards at secondary school's Mathematics Competitions. I was admitted to the High School Gifted Student , a top high school for talented pupils in Vietnam in 1994. My school is famous for having a number of academic awards achieved by students in national and international competitions in mathematics, informatics, chemistry, biology and physic. I started learning computer programming there (in Pascal) and had attended several programming competitions at the school, which is in similar format with ACM programming contests I got Bachelor of Engineering degree in Electronic and Telecommunication from Hanoi University of Technology , one of the leading universities in technology in Vietnam. My undergraduate study gave me strong back-ground on electronic design, programming, and telecommunication research especially in Robotics, Computer Vision, and Signal and Image Processing. During my undergraduate, I received annual scholarship for talented students as well as several other scholarship and fellowship for the top university students. I had also won a 2rd place at the school programming contest in 1999 and became a member of the school team at the national programming contest between top universities in Vietnam. I'm in the top undergraduate students and in the dean list honor. After graduate from the school, I worked as a research and teaching assistant at College of Technology, Vietnam National University in Hanoi. At the same time I got Master Degree in Electronic and Telecommunication there. In 2004, I were interviewed by top scientists from US and were awarded Vietnamese Education Foundation fellowship, I applied to the School of Computing University of Utah for my graduate study in the fall 2005. I took Computer Graphic class in the first semester at UoU, and immediately felt in love with the beauty of Graphic and Scientific Visualization, the research area in which University of Utah is one of the leading school in US. I changed my major to Computing Track. I joined Scientific Computing and Imaging Institute, one of the top US institutes in Visualization in spring 2006 and worked with Claudio T Silva until now. |
|
| Research Interests | |
|
I would like to challenge myself to discover what are my potential research abilities. I could do both theoretical (math and algorithms) and practical (electrical engineering, robotic, computer programming) research but I prefer practical ones. Recently, my main interest is in high performance computing (HPC) especially with general programming on GPUs (GPGPU). I found that there are many amazing things can be done with an enormous amount of computational power from GPUs. I like optimize performance of a program and try to improve myself to become a professional software developer. My favorite books are the Effective C++ programming series by Scott Meyers and Exception C++ series by Herb Sutter. |
|
| Thesis and Defense | |
|
High Performance Multi-scale image processing framework on
multiGPUs with applications to unbiased diffeomorphic atlas
construction
PhD thesis Defense presentation, February 2011. Committee members: Jens Kruger, Thomas Fletcher, Joao Comba, Sarang Joshi and Claudio T. Silva In the dissertation, I present: (1) essential components of a general 3D image processing library: data structures and algorithms, as well as how to implement these building blocks on the GPU architecture for optimal performance; (2) an implementation of unbiased atlas construction algorithms-an illustration of how to solve a highly complex and computationally expensive algorithm using this framework; (3) an extension of the framework to account for geometry descriptors to solve registration challenges with large scale shape changes and high intensity-contrast differences; and (4) an out-of-core streaming model, which enables developers to implement multi-image processing techniques on commodity hardware |
|
| Publications | |
 |
Optimal Multi-Image Processing Streaming Framework on Parallel
Heterogeneous Systems
Eurographic Symposium on Parallel Graphic and Visualization 2011 Linh Ha, Jens Kruger, Joao Comba, Sarang Joshi and Claudio T. Silva Best paper award at EGPGV 2011 Presentation The construction of a brain atlas often requires applying image processing operations to multiple images which is challenging due to large amount of computational and memory the construction requires. We introduce MIP, a Multi-Image Processing streaming framework to harness the processing power of heterogeneous CPU/GPU systems. We show specially designed streaming algorithms and data structures that provides an optimal solution for out-of-core multi-image processing problems both in terms of memory usage and computational efficiency. |
 |
Image Registration Driven by Combined Probabilistic and
Geometric Descriptors
International Conference on Medical Image Computing and Computer Assisted Intervention Linh Ha, Marcel Prastawa, Guido Gerig, John H. Gilmore, Claudio T. Silva and Sarang Joshi Deformable image registration in the presence of considerable contrast differences and large-scale size and shape changes represents a significant challenge for image registration. We propose a new registration method that generates a mapping between brain anatomies represented as a multi-compartment model of tissue class posterior images and geometries. |
 |
Multivariate Statistical Analysis of Deformation
Momenta Relating Anatomical Shape to Neuropsychological
Measures International Conference on Medical Image Computing and Computer Assisted Intervention Nikhil Singh, P. Thomas Fletcher, J. Samuel Preston, Linh Ha, Richard King, J. Stephen Marron, Michael Wiener and Sarang Joshi The purpose of this study is to characterize the neuroanatomical variations observed in neurological disorders such as dementia. We do a global statistical analysis of brain anatomy and identify the relevant shape deformation patterns that explain corresponding variations in clinical neuropsychological measures. |
 |
Multi-scale Unbiased Diffeomorphic Atlas
Construction on Multi-GPUs GPU Computing Gems Volume I L. Ha, Jens Kruger, Sarang Joshi and Claudio T. Silva In this chapter, we present a high performance multi-scale 3D image processing framework to exploit the parallel processing power of multiple graphic processing units (Multi-GPUs) for medical image analysis. We developed GPU algorithms and data structures that can be applied to a wide range of 3D image processing applications and efficiently exploit the computational power and massive bandwidth offered by modern GPUs. |
 |
Implicit radix sorting on GPUs GPU Computing Gems Volume II L. Ha, Jens Kruger and Claudio T. Silva In this chapter, we present a high performance sorting function on GPUs that is able to exploit the parallel processing power and memory bandwidth of modern GPUs to sort large quantities of data at a very high speed. We revisit the traditional radix sorting framework, analyze the weaknesses, and then propose a solution based on the implicit counting data presentation and its associated operations |
 |
Fast Parallel Unbiased Diffeomorphic Atlas
Construction on Multi-Graphics Processing Units Eurographic Symposium on Parallel Graphic and Visualization L. Ha, Jens Kruger and Claudio T. Silva This paper presents an efficient implementation of unbiased diffeomorphic atlas construction on the new parallel processing architecture based on Multi-Graphics Processing Units (Multi-GPUs). Our results show that the GPU implementation gives a substantial performance gain on the order of twenty to sixty times faster than a single CPU and provides an inexpensive alternative to large distributed-memory CPU clusters. |
 |
Fast 4-way parallel radix sorting on
GPUs Computer Graphics Forum L. Ha, Jens Kruger and Claudio T. Silva In this paper we present a hardware-optimized parallel implementation of the radix sort algorithm that results in a significant speed up over existing sorting implementations. We outperform all known GPU based sorting systems by about a factor of two and eliminate restrictions on the sorting key space. This makes our algorithm not only the fastest, but also the first general GPU sorting solution. |
 |
Spline-based Feature Curves from
Point-Sampled Geometry The Visual Computer 2008 Joel Daniels II, Tilo Ochotta, Linh Ha, and Claudio T. Silva An extension of the previous feature curve extraction method, leveraging robust moving least squares to identify and project points to sharp features. An iterative spline fitting method is used to smooth the projection-based noise while aligning cubic curves along the feature regions. |
|
Robust
Smooth Feature Extraction from Point Clouds Shape and Modeling International 2007 Joel Daniels II, Linh Ha, Tilo Ochotta and Claudio T. Silva A robust algorithm to extract features by defining polylines along the sharp edges of point-based models. Leverages the techniques of robust moving least squares to project points towards such regions. |
| Personal | |
|
I love cooking, playing sport, and listening to music in my free time. My favorite sports are swimming, skiing, badminton and soccer. I like traveling and taking a lot of pictures every where I come. I love the beauty of nature and human. Linh K. Ha |
|