Towards A Fairer Landmark Recognition Dataset
Abstract
We introduce a new landmark recognition dataset, which is created with a focus on fair worldwide representation. While previous work proposes to collect as many images as possible from web repositories, we instead argue that such approaches can lead to biased data. To create a more comprehensive and equitable dataset, we start by defining the fair relevance of a landmark to the world population. These relevances are estimated by combining anonymized Google Maps user contribution statistics with the contributors' demographic information. We present a stratification approach and analysis which leads to a much fairer coverage of the world, compared to existing datasets. The resulting datasets are used to evaluate computer vision models as part of the the Google Landmark Recognition and RetrievalChallenges 2021.
Cite
@article{arxiv.2108.08874,
title = {Towards A Fairer Landmark Recognition Dataset},
author = {Zu Kim and André Araujo and Bingyi Cao and Cam Askew and Jack Sim and Mike Green and N'Mah Fodiatu Yilla and Tobias Weyand},
journal= {arXiv preprint arXiv:2108.08874},
year = {2022}
}
Comments
Please cite the full detailed version of the paper instead: Improving Fairness in Large-Scale Object Recognition by CrowdSourced Demographic Information arXiv:2206.01326