Formlets

About the Formlet Representation of Planar Shape

The formlet representation is a sparse, multi-scale generative model of planar shape first presented at CVPR 2010[2], and then published in the Journal of Image and Vision Computing (JIVC) as an editor’s choice paper[1]. This page provides supplementary material for the JIVC manuscript.
We include our dataset, which consists of 391 blue-screened canonical animal images from the Hemera object database. The images are greyscale *.gif format. We provide a MATLAB implementation of the the Formlet Pursuit algorithm, as well as the Shapelet Pursuit algorithm proposed by Dubinskiy and Zhu [3,4]. We also provide code for extracting the parameter distributions and experimenting with various Formlet bases needed to reproduce the figures in the JIVC paper. To begin, please take a look at readme.txt.

If you have any questions, comments or bug reports, please contact Alex Yakubovich via yakuboa@yorku.ca.

References

[1] Oleskiw, T.D., Elder, J.H. & Peyré, G. (2010) On growth and formlets: sparse multi-scale coding of planar shape, in Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, Los Alamitos, CA, 459-466. (PDF)

[2] Elder, J.H., Oleskiw, T.D., Yakubovich, A. & Peyré, G. (2013). On growth and formlets: Sparse multi-scale coding of planar shape. Image and Vision Computing vol. 31, 1-13 (Editor’s Choice Paper). (PDF)

[3] Dubinskiy, A., Zhu, S.C.,”A multi-scale generative model for animate shapes and parts,” Computer Vision, 2003. Proceedings. Ninth IEEE International Conference on, vol. 1, pp. 249-256, Oct. 2003 (PDF)

[4] Dubinskiy, A., “Multi-scale generative model for animate contour representation: A Thesis”, Ohio State University, Oct. 2003

[5] Elder, J.H. (2013). Perceptual organization of shape. In S. Dickinson & Z. Pizlo, ed., Shape Perception in Human and Computer Vision: An Interdisciplinary Perspective, Springer. (PDF)

[6] Grenander, Ulf, Anuj Srivastava, and Sanjay Saini. “A pattern-theoretic characterization of biological growth.” Medical Imaging, IEEE Transactions on 26.5 (2007): 648-659. (PDF)

Downloads

Readme readme.txt
391 Natural Image Dataset 391animals.zip
Occluded Pursuit Experiment Occluded_Pursuit_Experiment.zip
Parameter Distributions Parameter_Distributions.zip
Selecting_Formlet_Basis Selecting_Formlet_Basis.zip