Research Project & Goals
My facet of research comes under the umbrella of a much larger, broader network as defined by Boston University's video
sensor network research group,
Sensorium. The Sensorium Infrastructure and associated projects in the Computer Science Department at
Boston University aim to catalyze fundamental advances in image and video computing, network protocols, and resource management
to deal with unique spatio-temporal constraints of sensor networks in general and of video sensor networks in particular.
When fully acquired, the Sensorium research infrastructure will be composed of a sensor network of video cameras spanning several rooms,
networked processing units, and a terabyte database, managed together to satisfy queries using those generated by mobile users within this environment.
The Sensorium infrastructure will enable a number of collaborative research projects led by various research groups in the department.
The following is a composite of mine and Jessie's contributions to the group:
Outlined research goals are as follows (to be updated as necessary):
- Get my 8 billion computer/research accounts up and running
- Read research papers and a thesis paper pertaining to computer vision, techniques, and human computer interaction programs
- Become familiar with the various research projects being worked on by graduate students
- Complete an introductory image processing assignment for a program that can detect skin color
- Finish installing the OpenCV library and understand its basic features
- Solve Linux compatibility issue pertaining the Logitech 4000 webcam driver
- Become familiar with standard computer vision techniques (motion detection, thresholding, segmentation...)
- Utilize primitive vision functions to build more compound functions
- Assist graduate students in their areas of research in computer vision
- Small programming tasks for other related image and video computing projects (eye tracking, Counting Fingers program, etc.)
- Establish benchmark testing and criteria for various still images and video (varying resolutions and other parameters)
- Utilize available C++ timing code (in milliseconds) or assembly code (in clock cycles) to determine the processing time of various computer vision functions given on the benchmark testing suite
- Build a table of timing values for both primitive and compound functions
- Analyze results and determine if additional research should be performed
(...with more to come later!)
|