9 - 11 September 2003
National Wine Centre, Adelaide, Australia
Organized by Michael J. Brooks and Kenichi
Kanatani
Sponsored by
The Japan Society for the Promotion of Science
The Australian Government's Innovation Access
Program
Tuesday 9 September 2003
Session 1: Factorization and Affine Reconstruction
- 9:40 -- 10:20
-
PowerFactorization: 3D reconstruction with missing or uncertain data,
Richard Hartley and Frederik Schaffalitzky (Australian National
University, Australia)
- 10:40 -- 11:20
-
Robust factorization for the affine camera: Analysis and comparison,
Du Q. Huynh* and Anders Heyden+
(*University of Western Australia, Australia,
+Malmo University, Sweden)
- 11:20 -- 12:00
-
Extending interrupted feature point tracking for 3-D affine
reconstruction,
Yasuyuki Sugaya and Kenichi Kanatani (Okayama University, Japan)
- 2:00 -- 2:40
-
Multi-stage optimization for multi-body motion segmentation,
Kenichi Kanatani and Yasuyuki Sugaya (Okayama University, Japan)
Session 2: Correspondence
- 2:40 -- 3:20
-
Robust image matching preserving global consistency,
Yasushi Kanazawa* and Kenichi Kanatani+
(*Toyohashi University of Technology, Japan,
+Okayama University, Japan)
- 3:40 -- 4:20
-
Fast algorithms for stereo matching and motion estimation,
Changming Sun (CSIRO Mathematical and Information
Sciences, Australia)
Session 3: Parameter Estimation
- 4:20 -- 5:00
-
3D accuracy improvement from an image: Evaluation and viewpoint
dependency,
Keisuke Kinoshita (ATR Human Information Science Laboratories, Japan)
Wednesday 10 September 2003
- 9:00 -- 9:40
-
A constrained parameter estimation approach for some vision problems,
Wojciech Chojnacki, Michael J. Brooks, Anton van den Hengel and
Darren Gawley (University of Adelaide, Australia)
- 9:40 -- 10:20
-
Curve fitting: orthogonal distance regression, principal curves and
full maximum likelihood,
Garry Newsam (Defense Science and Technology Organization, Australia)
- 10:40 -- 11:20
-
Recent advances in video based motion estimation: Very robust
estimators and improved direct factorization methods,
David Suter, Pei Chen, and Hanzi Wang (Monash University, Australia)
- 11:20 -- 12:00
-
Motion parameter estimation from optical flow without nuisance
parameters,
Naoya Ohta (Gunma University, Japan)
Session 4: Video and Surveillance Techniques
- 2:00 -- 2:40
-
Detecting moving objects, ghosts, and shadows in video streams,
Rita Cucchinara*,
Costantio Grana*,
Massimo Piccardi+,
Andrea Prati*
(*Universita di Modena e Geggio Emilia, Italy,
+University of Technology at Sydney, Australia)
- 2:40 -- 3:20
-
Skin segmentation using color and edge information,
Son Lam Phung, Abdesselam Bouzerdoum, and Douglas Chai
(Edith Cowan University, Australia)
Session 5: Media Aesthetics
- 3:40 -- 4:20
-
Virtual exhibition of traditional dances by blending colors of
multiple images,
Takeshi Shakunaga*, Yasuhiro Mukaigawa+,
Ryo Yamane*, Daisuke Genda* and Yuji Kamon*
(*Okayama University, Japan, +University of Tsukuba, Japan)
- 4:20 -- 5:00
-
Extracting Movie Tempo:
Computational media aesthetics at work,
Brett Adams*,
Chitra Dori+, and Svetha Venkatesh *
(*Curtin University of Technology, Australia,
+IBM T. J. Watson Research Center, U.S.A.)
Thursday 11 September 2003
- 9:00 -- 9:40
-
Automatic construction of optimal statistical shape models,
Rhodri H. Davis*, Carole J. Twining+,
Tim F. Cootes+, and Chris J. Taylor+
(*University of Melbourne, Australia,
+University of Manchester, U.K.)
- 9:40 -- 10:20
-
Surface normals to surfaces via shapelets,
Peter Kovesi (University of Western Australia)
Session 6: Shape
- 10:40 -- 11:20
-
Statistical methods for human behaviour recognition,
Goeff West and Svetha Venkatesh
(Curtin University of Technology, Australia)
Session 7: Multi-camera Geometry
- 11:20 -- 12:00
-
Recovering epipolar geometry from mutual projections of multiple
cameras,
Jun Sato (Nagoya Institute of Technology, Japan)
- 2:00 -- 2:40
-
Self-calibration from two perspective views under various conditions:
Closed-form solutions and degenerate configurations,
Toshio Ueshiba and Fumiaki Tomita (National Institute of Advanced
Industrial Science and Technology, Japan)
- 2:40 -- 3:20
-
Diverging viewing-lines in binocular vision: A method for estimating
ego motion by mounted active cameras,
Akihiro Sugimoto* and Tomohiko Ikeda+
(*National Institute of Informatics, Japan,
+Chiba University, Japan).
- 3:20 -- 4:00
-
Construction of 3D map in robot by combining action and perceived
images,
Koichiro Deguchi (Tohoku University, Japan)