Applying random forests to the problem of dense non-rigid shape correspondence

Vestner M, Rodola E, Windheuser T, Bulo SR, Cremers D (2016)

Publication Type: Book chapter / Article in edited volumes

Publication year: 2016

Publisher: Springer Heidelberg

Edited Volumes: Visualization in Medicine and Life Sciences III - Towards Making an Impact

Series: Mathematics and Visualization

Pages Range: 231-248

DOI: 10.1007/978-3-319-24726-7_11

Abstract

We introduce a novel dense shape matching method for deformable, three-dimensional shapes. Differently from most existing techniques, our approach is general in that it allows the shapes to undergo deformations that are far from being isometric. We do this in a supervised learning framework which makes use of training data as represented by a small set of example shapes. From this set, we learn an implicit representation of a shape descriptor capturing the variability of the deformations in the given class. The learning paradigm we choose for this task is a random forest classifier. With the additional help of a spatial regularizer, the proposed method achieves significant improvements over the baseline approach and obtains state-of-the-art results while keeping a low computational cost.

Involved external institutions

Technische Universität München (TUM)

Germany (DE) Fondazione Bruno Kessler (FBK) (früher: ITC-irst)

Italy (IT)

How to cite

APA:

Vestner, M., Rodola, E., Windheuser, T., Bulo, S.R., & Cremers, D. (2016). Applying random forests to the problem of dense non-rigid shape correspondence. In Lars Linsen, Hans-Christian Hege, Bernd Hamann (Eds.), Visualization in Medicine and Life Sciences III - Towards Making an Impact. (pp. 231-248). Springer Heidelberg.

MLA:

Vestner, Matthias, et al. "Applying random forests to the problem of dense non-rigid shape correspondence." Visualization in Medicine and Life Sciences III - Towards Making an Impact. Ed. Lars Linsen, Hans-Christian Hege, Bernd Hamann, Springer Heidelberg, 2016. 231-248.

BibTeX: Download