Color-Based Object Recognition by a Grid-Connected Robot Dog
Overview of the Demonstration:
This demonstration shows object recognition performed by a Sony Aibo
robot dog. The dog is connected to a wide-area Grid system, potentially
consisting of hundreds of computers located at several institutes in
Europe, the United States, and Australia. Apart from the quality of
recognition, we demonstrate the effectiveness of Grid usage in
multimedia computing.
Moreover, we show the ease with which multimedia applications can be
integrated with Grid computing.
What we have to offer:
|
  |
Short movie
showing our Aibo robot dog walking around, and recognizing
the objects it has learned.
|
|   |
  |
  |
|
  |
Promotional video
"Color-Based Object Recognition by a Grid-Connected Robot Dog",
which has won a most visionary research award at
AAAI-07, and gained an "honourable mention"
at the annual Dutch meeting between science and press: Bessensap
2007. see also the
"Noorderlicht" pages of the Dutch broadcasting cooperation
VPRO.
|
|
|   |
  |
  |
|
  |
Full image
showing our World-Wide Aibo-Grid initialization GUI, which allows
us to select computing resources worldwide with only a few
mouse-clicks.
|
|   |
  |
  |
|
  |
Short movie
showing the Aibo Object Recognition GUI, in which the 'learning'
phase is performed for one object, while being connected to four
cluster computers in Europe and one in Australia.
|
|   |
  |
  |
|
  |
Short movie
showing about 50 out of
1,000 objects
that our Aibo robot dog has learned.
|
|   |
  |
  |
|
  |
Pictures
taken during demonstration sessions at international
conferences and national ICT meetings.
|
|   |
  |
  |
|
  |
Picture shot during the
making of the promotional
video "Color-Based Object Recognition by a Grid-Connected
Robot Dog", Amsterdam, The Netherlands, April 18, 2006.
|
|   |
  |
  |
|
  |
Poster
used for our demonstration at ICME 2005, Amsterdam,
The Netherlands, July 6-8, 2005.
|
|   |
  |
  |
|
  |
Poster
used for our demonstrations at ECCV 2006 (Graz, Austria, May 7-13,
2006) and Supercomputing 2007 (Reno, NV, USA, November 10-16, 2007)
- with a significant increase in the available resources since ICME
2005.
|
People:
Acknowledgements:
We would like to thank the following people for their support,
and for granting us access to their cluster systems:
- Australia
- Prof. David Abramson, Colin Enticott, Pim Amponpun (Monash
University, Melbourne)
- Prof. Lindsay Botten, Achim Casties, Vladas Leonas (Australian
Centre for Advanced Computing and Communications, Sydney-Eveleigh)
- Prof. Albert Zomaya, Philip McCrea, Chen Wang (University of
Sydney)
- Tony Adriaansen (CSIRO, Sydney)
- United States
- Bhiksha Raj, Ron Johnson (Mitsubishi Electric Research
Laboratories, Boston - Cambridge, MA)
- Jason Leigh, Lance Long, Alan Verlo (University of Illinois at Chicago, IL)
- Philip Papadopoulos (San Diego Supercomputer Center, University
of California, San Diego, CA)
- Europe
- Andreas Uhl, Ernst Forsthofer (Salzburg University, Salzburg,
Austria)
- Tobias Klug, Carsten Trinitis (Technical University Munich,
Germany)
- Daniele D'Agostino, Antonella Galizia (University of Genoa, Italy)
- Marian Bubak, Patryk Lason, Lukasz Skital (University of Science
and Technology, Krakow, Poland)
- Norbert Meyer, Piotr Siwczak, Marek Zawadzki (Poznan Supercomputing
& Networking Center, Poznan, Poland)
- Jesus Marco de Lucas, Rafael Marco de Lucas (Instituto de Fisica
de Cantabria, Santander, Spain)
- Prof. Andrew Zisserman (Oxford University, United Kingdom)
- The Netherlands
- Prof. Henri Bal, Thilo Kielmann, Andre Merzky, Kees Verstoep
(Vrije Universiteit, Amsterdam)
- Prof. Cees de Laat (Universiteit van Amsterdam)
We are grateful to Michiel van Liempt (Universiteit van Amsterdam) for
his excellent efforts in implementing the original sequential object
recognition code.
Finally, thanks go out to Edwin Steffens and Arnoud Visser for providing
us with a Sony Aibo robot dog.
|