Greedy Structure Learning of Hierarchical Compositional Models

Adam Kortylewski, Aleksander Wieczorek, Mario Wieser, Clemens Blumer, Sonali Parbhoo, Andreas Morel-Forster, Volker Roth, Thomas Vetter; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 11612-11621

Abstract


In this work, we consider the problem of learning a hierarchical generative model of an object from a set of images which show examples of the object in the presence of variable background clutter. Existing approaches to this problem are limited by making strong a-priori assumptions about the object's geometric structure and require seg- mented training data for learning. In this paper, we propose a novel framework for learning hierarchical compositional models (HCMs) which do not suffer from the mentioned limitations. We present a generalized formulation of HCMs and describe a greedy structure learning framework that consists of two phases: Bottom-up part learning and top-down model composition. Our framework integrates the foreground-background segmentation problem into the structure learning task via a background model. As a result, we can jointly optimize for the number of layers in the hierarchy, the number of parts per layer and a foreground- background segmentation based on class labels only. We show that the learned HCMs are semantically meaningful and achieve competitive results when compared to other generative object models at object classification on a standard transfer learning dataset.

Related Material


[pdf]
[bibtex]
@InProceedings{Kortylewski_2019_CVPR,
author = {Kortylewski, Adam and Wieczorek, Aleksander and Wieser, Mario and Blumer, Clemens and Parbhoo, Sonali and Morel-Forster, Andreas and Roth, Volker and Vetter, Thomas},
title = {Greedy Structure Learning of Hierarchical Compositional Models},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}