CLIP Augmentation for Image Search

Ingus Pretkalnins, Arturs Sprogis, Guntis Barzdins

2022

Abstract

We devised a probabilistic method for adding face recognition to the neural network model CLIP. The method was tested by creating a prototype and matching 1000 images to their descriptions. The method improved the text to image Recall @ 1 metric from 14.0% matches for CLIP alone to 21.8% for CLIP + method, for a sample size of 1000 images and descriptions.

Paper Citation

in Harvard Style

Pretkalnins I., Sprogis A. and Barzdins G. (2022). CLIP Augmentation for Image Search. In Proceedings of the 7th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS, ISBN 978-989-758-565-4, pages 71-78. DOI: 10.5220/0011046600003197

in Bibtex Style

@conference{complexis22,
author={Ingus Pretkalnins and Arturs Sprogis and Guntis Barzdins},
title={CLIP Augmentation for Image Search},
booktitle={Proceedings of the 7th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS,},
year={2022},
pages={71-78},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011046600003197},
isbn={978-989-758-565-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Complexity, Future Information Systems and Risk - Volume 1: COMPLEXIS,
TI - CLIP Augmentation for Image Search
SN - 978-989-758-565-4
AU - Pretkalnins I.
AU - Sprogis A.
AU - Barzdins G.
PY - 2022
SP - 71
EP - 78
DO - 10.5220/0011046600003197