This work introduces composed image retrieval to remote sensing. It allows to query a large image archive by image examples alternated by a textual description, enriching the descriptive power over unimodal queries, either visual or textual. Various attributes can be modified by the textual part, such as shape, color, or context. A novel method fusing image-to-image and text-to-image similarity is introduced. We demonstrate that a vision-language model possesses sufficient descriptive power and no further learning step or training data are necessary. We present a new evaluation benchmark focused on color, context, density, existence, quantity, and shape modifications. Our work not only sets the state-of-the-art for this task, but also serves as a foundational step in addressing a gap in the field of remote sensing image retrieval. Code at: https://github.com/billpsomas/rscir
@article{arxiv.2405.15587,
title = {Composed Image Retrieval for Remote Sensing},
author = {Bill Psomas and Ioannis Kakogeorgiou and Nikos Efthymiadis and Giorgos Tolias and Ondrej Chum and Yannis Avrithis and Konstantinos Karantzalos},
journal= {arXiv preprint arXiv:2405.15587},
year = {2024}
}
Comments
Accepted for ORAL presentation at the 2024 IEEE International Geoscience and Remote Sensing Symposium