Hierarchical Deformable Part Models for Heads and Tails

Fatemeh Shokrollahi Yancheshmeh, Ke Chen, Joni-Kristian Kämäräinen

Abstract

Imbalanced long-tail distributions of visual class examples inhibit accurate visual detection, which is addressed by a novel Hierarchical Deformable Part Model (HDPM). HDPM constructs a sub-category hierarchy by alternating bootstrapping and Visual Similarity Network (VSN) based discovery of head and tail sub-categories. We experimentally evaluate HDPM and compare with other sub-category aware visual detection methods with a moderate size dataset (Pascal VOC 2007), and demonstrate its scalability to a large scale dataset (ILSVRC 2014 Detection Task). The proposed HDPM consistently achieves significant performance improvement in both experiments.

Download


Paper Citation


in Harvard Style

Yancheshmeh F., Chen K. and Kämäräinen J. (2018). Hierarchical Deformable Part Models for Heads and Tails.In Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, ISBN 978-989-758-290-5, pages 45-55. DOI: 10.5220/0006532700450055


in Bibtex Style

@conference{visapp18,
author={Fatemeh Shokrollahi Yancheshmeh and Ke Chen and Joni-Kristian Kämäräinen},
title={Hierarchical Deformable Part Models for Heads and Tails},
booktitle={Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP,},
year={2018},
pages={45-55},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006532700450055},
isbn={978-989-758-290-5},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP,
TI - Hierarchical Deformable Part Models for Heads and Tails
SN - 978-989-758-290-5
AU - Yancheshmeh F.
AU - Chen K.
AU - Kämäräinen J.
PY - 2018
SP - 45
EP - 55
DO - 10.5220/0006532700450055