>> Localized Components Analysis
Two localized vectors affecting the jaw in different ways
Two localized vectors. Both affect the jaw region in different ways.

Abstract

Poster We introduce Localized Components Analysis (LoCA) for describing surface shape variation in an ensemble of biomedical objects using a linear subspace of spatially localized shape components. In contrast to earlier methods, LoCA optimizes explicitly for localized components and allows a flexible trade-off between localized and concise representations. Experiments comparing LoCA to a variety of competing shape representation methods on 2D and 3D shape ensembles establish the superior ability of LoCA to modulate the locality-conciseness tradeoff and generate shape components corresponding to intuitive modes of shape variation. Our formulation of locality in terms of compatibility between pairs of surface points is shown to be flexible enough to enable spatially-localized shape descriptions with attractive higher-order properties such as spatial symmetry.

Results

Comparison between PCA and LoCA axes

LoCA was performed on several data sets. The summary images show the 16 basis vectors minimizing the reconstruction error, where each vector is represented by an object/graph pair. The object represents the average item of the data set, and the arrows indicate the degree that the object changes as the corresponding shape parameter is varied.

The graphs plot the magnitude of basis vector's effect on a point against the point's distance to the vector's "center point". The center point of the region is defined as the point that minimizes our cost function.

The example on the right corresponds to the animation above. One can see that PCA generates a global deformation, bending the corpus callosum in half. In contrast, LoCA generates a localized deformation that expands the genu and leaves everything else alone.

The accompanying movies show how the shape parameters alter objects in the space. Each parameter is controlled by a slider, and is initially set to 0. The ends of each slider are capped at 4 standard deviations away from 0. Each slider doubles as a histogram showing how many objects from the original data set have a particular value for the corresponding parameter.

Corpora Callosa

55 2D tracings of corpora callosa were used, each consisting of 103 points.

Principal Components Analysis
7/54 vectors required to get 90% variation
Localized Components Analysis
Parameters set so 26/54 vectors required to get 90% variation
PCA: First 16 corpora callosa basis vectors
First basis vectors (Large image) (Movie)
LoCA: First 16 corpora callosa basis vectors
First basis vectors (Large image) (Movie)

Monkey crania

A set of 239 monkey crania from different species were used. Each cranium was represented by a set of 45 points.
Principal Components Analysis
26/135 vectors required to get 90% variation
Localized Components Analysis
Parameters set so that 82/135 vectors were required to get 90% variation
PCA: First 16 crania basis vectors, front view
Front view of the first vectors (Large image) (Movie)
LoCA: First 16 crania basis vectors, front view
Front view of the first vectors (Large image) (Movie)
PCA: First 16 crania basis vectors, side view
Side view of the first vectors (Large image) (Movie)
LoCA: First 16 crania basis vectors, side view
Side view of the first vectors (Large image) (Movie)
Symmetric Localized Components Analysis
In addition to distance, symmetry is taken into account to determine compatibility. Parameters were set so that 64/135 vectors were required for 90% variation
Symmetric LoCA: First 16 crania basis vectors, front view
First basis vectors (front view) (Large image) (Movie)
Symmetric LoCA: First 16 crania basis vectors, side view
First 16 basis vectors (side view) (Large image) (Movie)
LoCA: First 16 crania basis vectors, under view
First 16 basis vectors (under) (Large image) (Movie)

Lateral ventricles

A set of 54 lateral ventricles from humans with and without AIDS were used. Each ventricle was represented by a set of 300 points.
Principal Components Analysis
17/54 vectors required to get 90% variation
Symmetric Localized Components Analysis
Parameters set so that 31/54 vectors were required to get 90% variation
PCA: First 16 ventricle basis vectors
LoCA: First 16 ventricle vectors

Acknowledgments

This research was funded by NSF gransts IIS--0513894 and IIS--0513660.

Publications

  • D. Alcantara, O. Carmichael, E. Delson, W. Harcourt-Smith, K. Sterner, S. Frost, R. Dutton, P. Thompson, H. Aizenstein, O. Lopez, J. Becker, and N. Amenta, "Localized Components Analysis", Proceedings of IPMI, 2007, pp. 519 - 531
  • D. Alcantara, O. Carmichael, E. Delson, W. Harcourt-Smith, K. Sterner, S. Frost, R. Dutton, P. Thompson, and N. Amenta, "Exploration of Shape Variation Using Localized Components Analysis", IEEE Transactions on Pattern Analasysis and Machine Intelligence, (to appear, 2009)