Converting Image Labels to Meaningful and Information-rich Embeddings

Savvas Karatsiolis, Andreas Kamilaris

2021

Abstract

A challenge of the computer vision community is to understand the semantics of an image that will allow for higher quality image generation based on existing high-level features and better analysis of (semi-) labeled datasets. Categorical labels aggregate a huge amount of information into a binary value which conceals valuable high-level concepts from the Machine Learning models. Towards addressing this challenge, this paper introduces a method, called Occlusion-based Latent Representations (OLR), for converting image labels to meaningful representations that capture a significant amount of data semantics. Besides being information-rich, these representations compose a disentangled low-dimensional latent space where each image label is encoded into a separate vector. We evaluate the quality of these representations in a series of experiments whose results suggest that the proposed model can capture data concepts and discover data interrelations.

Paper Citation

in Harvard Style

Karatsiolis S. and Kamilaris A. (2021). Converting Image Labels to Meaningful and Information-rich Embeddings.In Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-486-2, pages 107-119. DOI: 10.5220/0010375801070119

in Bibtex Style

@conference{icpram21,
author={Savvas Karatsiolis and Andreas Kamilaris},
title={Converting Image Labels to Meaningful and Information-rich Embeddings},
booktitle={Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2021},
pages={107-119},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010375801070119},
isbn={978-989-758-486-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Converting Image Labels to Meaningful and Information-rich Embeddings
SN - 978-989-758-486-2
AU - Karatsiolis S.
AU - Kamilaris A.
PY - 2021
SP - 107
EP - 119
DO - 10.5220/0010375801070119