Research Interests

Jim Holten

Home page	Background	Research	Publications	Courses

Applying complex systems analysis in a wide variety of application disciplines
- Modeling Critical Infrastructure Networks
- Automated Parallel Decomposition
- Taming IT (Information Technologies) via the SRF Algebra
- Social and Critical Infrastructure Networks Modeling

Computer and Robot Vision

Some of the difficulties in computer vision that have delayed the development of robust computer vision systems include:

Left Camera View -------	------- Right Camera View

-Note the differences in camera image quality to be overcome in matching between images. -Also note the "nose" which helps in the automation of camera calibration for angles to 3-D distances, but only if it can be "registered" between images. -The overall "clutter" is one of the more difficult aspects for computer recognition of real scene objects.

Outdoor Scene Sequence (real clutter, pixel color vectors) -------

In Greyscale (simple pixel value) -------

"Smoothness" vs "Edgeness" -------

Some additional goals of stereo vision include:

Stereo and image sequence 3-D environment modeling, including error characterization
Vision for robot path determination
Modeling visual object dynamics and appearances for obstacle avoidance and object characterization
Object recognition for robot lanndmark navigation and target acquisition and tracking
Image content understanding

Spinoffs from computer vision --

Real error tracking and control in awkward and large-scale numeric techniques
Generalized characterization and comparison of the similarity of objects with nonhomogeneous and inexact attributes
Extreme performance hardware architectures for AI, vision, speech understanding, and other specialized algorithms (i.e. GPUs, FPGAs, multicore exploitation)
Asynchronous algorithm design and development techniques.

Intelligent Machines/Robots/Artificial Intelligence
- Cerebral cortex and mental processes modeling
- Concept nets for multiple written and spoken language understanding and translation
- Actor/Event models
- Relating observed information to "meaningful" processes, interpretations, and forming alternative predictive expectations
- Formulating and relating multiple levels of abstractions and interpretive contexts
Taming Informatics via a Common Low-level Abstraction
Spinoffs from Taming Informatics --
- Beyond data structure basics: ADAM, SRF, n-D mesh representations, and attributed graphs
- Parallel and Distributed Data Computing
  - Parallel I/O
  - Dynamic Repartitioning
  - Fault-tolerant Large Software System Design
- Sequential and parallel (mixed) algorithms and complex data structures partitioning
- Computer Model Types and Interoperability
  - Generalized Mesh Models
  - Finite Element Models
  - CAD Models
  - Social and Critical Infrastructures Network Models
- Protein sequences to 3-D configurations
Going beyond CMM and agile methods: Adaptive software engineering approaches
- Sequential Code via Associative Techniques -- storyboarding
- The power of self-discipline for generating quality code -- avoiding bad design and coding habits
Computer science education
- Growing beyond language mechanics
- The science behind the technologies
- The engineering based on the science and technologies
- The art of applying the science and engineering

Other Links: