Graph Navigation for Exploring Very Large Image Collections

Kai Uwe Barthel, Nico Hezel

2017

Abstract

We present a new approach to visually browse very large sets of untagged images. In this paper we describe how to generate high quality image descriptors/features using transformed activations of a convolutional neural network. These features are used to model image similarities, which again are used to build a hierarchical image graph. We show how such an image graph can be constructed efficiently. After investigating several browsing and visualization concepts, we found best user experience and ease of usage is achieved by projecting sub-graphs onto a regular 2D-image map. This allows users to explore the image graph similar to navigation services.

Paper Citation

in Harvard Style

Barthel K. and Hezel N. (2017). Graph Navigation for Exploring Very Large Image Collections . In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017) ISBN 978-989-758-226-4, pages 411-416. DOI: 10.5220/0006274804110416

in Bibtex Style

@conference{visapp17,
author={Kai Uwe Barthel and Nico Hezel},
title={Graph Navigation for Exploring Very Large Image Collections},
booktitle={Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017)},
year={2017},
pages={411-416},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006274804110416},
isbn={978-989-758-226-4},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017)
TI - Graph Navigation for Exploring Very Large Image Collections
SN - 978-989-758-226-4
AU - Barthel K.
AU - Hezel N.
PY - 2017
SP - 411
EP - 416
DO - 10.5220/0006274804110416