Metric Learning for Image Alignment
People
Abstract
Image alignment has been a long standing problem
in computer vision. Parameterized Appearance Models
(PAMs) such as the Lucas-Kanade method, Eigentracking,
and Active Appearance Models are commonly used to
align images with respect to a template or to a previously
learned model.While PAMs have numerous advantages relative
to alternate approaches, they have at least two drawbacks.
First, they are especially prone to local minima in the
registration process. Second, often few, if any, of the local
minima of the cost function correspond to acceptable solutions.
To overcome these problems, this paper proposes
a method to learn a metric for PAMs that explicitly optimizes
that local minima occur at and only at the places corresponding
to the correct fitting parameters. To the best of
our knowledge, this is the first paper to address the problem
of learning a metric to explicitly model local properties of
the PAMs error surface. Synthetic and real examples show
improvement in alignment performance in comparison with
traditional approaches. In addition, we show how the proposed
criteria for a good metric can be used to select good
features to track.
Citation
 |
Metric Learning for Image Alignment. Nguyen, M.H. and De la Torre, F. (2009) International Journal of Computer Vision, accepted.
[PDF] [Bibtex]
|
 |
Local Minima Free Parameterized Appearance Models. Nguyen, M.H. and De la Torre, F. (2008) Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.
[PDF] [Bibtex]
|
 |
Learning Image Alignment without Local Minima for Face Detection and Tracking. Nguyen, M.H. and De la Torre, F. (2008) Proceedings of the 8th IEEE International Conference on Automatic Face and Gesture Recognition.
[PDF] [Bibtex]
|
Talks
- Local Minima Free Parameterized Appearance Models, Machine Learning Lunch, Carnegie Mellon University.
Video
- Learning Image Alignment without Local Minima for Face Detection and Tracking, oral presentation, Face and Gesture 08. Slides
Acknowledgements and Funding
This work was supported by the US Naval Research Laboratory
under Contract no. N00173-07-C-2040. Any opinions, findings,
and conclusions or recommendations expressed in this material
are those of the author(s) and do not necessarily reflect the views
of the US Naval Research Laboratory.
Copyright notice