Projects

The project in this course represents a chance for you to develop and demonstrate your ability to do independent work. This is an opportunity, but also a responsibility and to take seriously and involves a significant time commitment. Start your work early. Plan a schedule and stick to it.

Deliverables

Proposal (15%)

Final project report (35%) In addition to above, describe in detail the work you did, describe the results (experimental and/or theoretical), and discuss their relevance and applicability. The format and quality of the report should be similar to a short (10-12 pages article style or 6-8 pages IEEE twocolumn) research paper. Study the style and conventions in the papers you read for the proposal.

Resources

Search resources and research journals in the library.
Searchable index of on-line articles by NEC Research Index
The original on-line paper search engine CORA

Suggested topics

Neurosci
  Georgopoulos, Kalaska, Shieber...

PercAct
  Goodale, Lomis, 

Characterizing motion:
  Activity recognition: Nelson, Alimonios
  Fleet, Jepson...
  Jagersand
  Matlab implementaton

Animating motion.
  
Tracking:
  Hager, Black, 
  XVision: practical experiments with.
  Filtering, Bar-Shalom:

Robotics: 
  Controller HW and SW for the small robot arm (or the PUMA's)

Real time systems
  Real time linux


------------------------------

The visual brain in action

Title: 
            The visual brain in action / A. David Milner and Melvyn A. Goodale. 
Author: 
            Milner, A. D. (A. David) 
            Goodale, Melvyn A. 

------------------------------

A more theoretical project in Projective geom:

Intro and tutorial:
===================
http://www.inrialpes.fr/movi/people/Triggs/isprs96/isprs96.html

Also borrow from lib:
3D computer vision
by Oliver Faugeras

Robotics/motion application:
============================
What Tasks Can Be Performed with an Uncalibrated Stereo Vision System? (with J. Hespanha, Z.Dodds, and A.S. Morse) Submitted for review to
IJCV. (21 pages, 300K compressed postscript) 
      Read the Abstract. 

See Greg's www page for postscript files:
http://www.cs.jhu.edu/~hager/Public/Publications/

Practical on robot control or user interfaces:
==============================================

Image Based Visual Simulation and Tele-Assisted Robot Control by J鋑ersand M. In IROS '97, Proc. New Trends in Image Based Visual
      Servoing. 


      Vision-Based Reach-To-Grasp Movements: From the Human Example to an Autonomous Robotic System by Alexa Hauck 
      A Tutorial Introduction to Visual Servo Control G. Hager, S. Hutchinson and P. Corke IEEE Transactions on Robotics and Automation, 12(5) pp.
      651-670, 1996. (122K compressed postscript) 

(See course WWW page)

Matlab:
robot toolbox
zebra toolbox
visfeedb toolbox: Visual control
http://ugweb.cs.ualberta.ca/~c306/docs/matlabdirs/libintro/libintro.ps 
http://ugweb.cs.ualberta.ca/~c306/docs/matlabdirs/

Doc: rob toolbox: P. Corke, look on net.

P Corke: "Visual servoing" book, about 94 or 96. Try Interlibrary

Zach's thesis on the course www page

Grosso, Sandini optic flow -> vis control


---------------------------------------------------------------------------




------------------------------------------
Zhen Deng

Vision based robot control

The material on the www site should be a good start.
As you have already noticed Alexa's thesis is perhaps the best written
one. Pick the chapters related to robot modelling.
For solving of whole tasks see my paper,
and finally for a bit of geometry challenge: take at least a brief 
look at Zach's excellent thesis.

Zach and Greg will come here this summer to visit.


-----------------------------------------------------
Cheryl Card
Something involving functional modelling of the human visual-motor system?

Alexa's: I recommend reading the human related chapters to get an idea
of it's relation to control as we think of it. Then if you have
time recurse back to the details about vision based control.

Milner, Goodale gives a much less technical, yet interesting
view of motion control.

For some current work on spatial transforms etc:

Vis-Mot trans in par cortex. Andersen lab
http://vis.caltech.edu/

-----------------------------------------------
Keith Yerex

Linear filter and subspace based approaches to capture, characterize
and reanimate lim motions

See the motion papers we read. In addition you can check:

Nayar, Nene Murase: Subspace methods for robot vision, 
Columbia CAVE TR about 1995. Is available on-line

From Animation 2000 conference. You saw my talk on this, right?
http://www.cs.ualberta.ca/~jag/publications/MotVisAnim2000.ps

----------------------------------------------------------

Chris Parker, Sean Verret

Dynamic models for tracking

For practical tracking:

The XVision System: A General-Purpose Substrate for Portable Real-Time Vision Applications (with K. Toyama). In Computer Vision and Image Understanding
69(1) pp. 23 - 37. (32 pages, 1.5M compressed postscript). 

Available on Greg's publication page:
  http://www.cs.jhu.edu/~hager/Public/Publications/

We will download the newest version of XVision and work with it.
See the CIPS link on the course www page for info.

For more fundamental stuff on filtering etc: Bar-Shalom and Fortman
"Tracking and Data Assocoation" Maybe you have your own books on
this also.

Check Andrew Blake and others work on tracking here:
http://www.robots.ox.ac.uk/
http://www.robots.ox.ac.uk/~misard/condensation.html