H. Quynh Dinh: Grants

Detecting Patterns in Vector Fields
H.Q. Dinh
Honda Initiation Grant, Honda Research Institute, 2009-2010.
Detecting patterns in vector fields will become increasingly crucial as computational methods for simulating fluid dynamics (CFD), sensor technology for dynamic data, and video surveillance increase in accuracy and ubiquity. The goal of this research is an in-depth study into how patterns can be detected in vector fields, starting with 2D fields and extending to fields over surfaces and in volumes (3D fields). Over the next 5 years, we will extend these algorithms to complex domains on simulated and recorded data via collaborations with researchers in automotive and ocean engineering and medical imaging.

Related publications:
  • L. Xu and H.Q. Dinh. "A Local Descriptor for Finding Corresponding Points in Vector Fields", IAPR International Conference on Pattern Recognition (ICPR), Dec. 2008 PDF 1.4MB
  • H.Q. Dinh and L. Xu. "Measuring the Similarity of Vector Fields Using Global Distributions", IAPR International Workshop on Structural and Syntactic Pattern Recognition (SSPR) in conjunction with ICPR, Dec. 2008 PDF 3.8MB
A Transderivational Search Engine for Creative Analogy Generation in Mixed-Media Design
H.Q. Dinh and E. Fisher
National Science Foundation Creative IT Program, Award# IIS-0742440, 2007-2009.
Text-based search engines are now reaching maturity. More recently, content-based (non-textual) retrieval algorithms have been developed for music, images, video, and 3D shapes. The goal of this project is to develop a transderivational search engine that suggests analogies across different media forms (e.g., text, audio, images, video, and 3D shapes) by looking at structural similarity within media content. The search engine will be developed in the context of designing interactive, mixed-media installations and in a brainstorming application for artists and designers. The result will be a transformative technology at the intersection of art, computer graphics, machine learning, cognitive psychology, and human-computer interaction (HCI). Project Webpage

Related publications:

  • V. Petite , H.Q. Dinh, and E. Fisher. "A User Study on Perceiving Analogies in Images", International Digital Media and Arts Association (iDMAa) Conference, 2008. PDF 177K
  • H.Q. Dinh and S. Kropac. "Multi-Resolution Spin-Images", Computer Vision and Pattern Recognition (CVPR), June 2006 PDF 1.3M
Automated Construction of Digital Models from Real Artifacts
H.Q. Dinh, G. Turk, G. Slabaugh, and R. Schafer
GVU Seed Grant, 2000
Quality graphics is a growing presence in the home due to the diminishing price of computer hardware. As computers become more prevalent, activities such as creating web pages, image editing, and video editing have become more commonplace and no longer require professional training. Creating digital models, however, require special equipment or training. The goal of this project is to develop automated methods that will enable non-professional, home PC users, and graphic art enthusiasts to create their own digital models from existing artifacts.

Related publications:
  • H.Q. Dinh, G. Turk, and G. Slabaugh. "Reconstructing Surfaces by Volumetric Regularization Using Radial Basis Functions", IEEE Transactions on Pattern Analysis and Machine Intelligence, October 2002 PDF 2.9M
  • H.Q. Dinh, G. Turk, and G. Slabaugh. "Reconstructing Surfaces Using Anisotropic Basis Functions", International Conference on Computer Vision (ICCV), 2001 PDF 1.3M

H. Quynh Dinh: Homepage