Exploitation of Noisy Automatic Data Annotation and Its Application to Hand Posture Classification

Georgios Lydakis, Georgios Lydakis, Iason Oikonomidis, Dimitrios Kosmopoulos, Antonis A. Argyros, Antonis A. Argyros

2022

Abstract

The success of deep learning in recent years relies on the availability of large amounts of accurately annotated training data. In this work, we investigate a technique for utilizing automatically annotated data in classification problems. Using a small number of manually annotated samples, and a large set of data that feature automatically created, noisy labels, our approach trains a Convolutional Neural Network (CNN) in an iterative manner. The automatic annotations are combined with the predictions of the network in order to gradually expand the training set. In order to evaluate the performance of the proposed approach, we apply it to the problem of hand posture recognition from RGB images. We compare the results of training a CNN classifier with and without the use of our technique. Our method yields a significant increase in average classification accuracy, and also decreases the deviation in class accuracies, thus indicating the validity and the usefulness of the proposed approach.

Paper Citation

in Harvard Style

Lydakis G., Oikonomidis I., Kosmopoulos D. and Argyros A. (2022). Exploitation of Noisy Automatic Data Annotation and Its Application to Hand Posture Classification. In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP; ISBN 978-989-758-555-5, SciTePress, pages 632-641. DOI: 10.5220/0010906200003124

in Bibtex Style

@conference{visapp22,
author={Georgios Lydakis and Iason Oikonomidis and Dimitrios Kosmopoulos and Antonis A. Argyros},
title={Exploitation of Noisy Automatic Data Annotation and Its Application to Hand Posture Classification},
booktitle={Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP},
year={2022},
pages={632-641},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010906200003124},
isbn={978-989-758-555-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2022) - Volume 5: VISAPP
TI - Exploitation of Noisy Automatic Data Annotation and Its Application to Hand Posture Classification
SN - 978-989-758-555-5
AU - Lydakis G.
AU - Oikonomidis I.
AU - Kosmopoulos D.
AU - Argyros A.
PY - 2022
SP - 632
EP - 641
DO - 10.5220/0010906200003124
PB - SciTePress