GeoDE:一个用于目标识别的地理多样性评估数据集
GeoDE: a Geographically Diverse Evaluation Dataset for Object Recognition
AI总结:
本文提出地理多样性目标识别评估数据集GeoDE,通过向全球征集图像构建,包含6个地区40类共61940张图片,用于评估和训练以揭示并缓解现有模型的偏见与缺陷。
摘要:
当前数据集收集方法通常从网络抓取大量数据。虽然这种技术极具可扩展性,但以此方式收集的数据往往会强化刻板偏见,可能包含个人身份信息,且通常源自欧洲和北美洲。在这项工作中,我们重新思考数据集收集范式,并引入 GeoDE,这是一个具有地理多样性的数据集,包含来自 40 个类别和 6 个世界地区的 61,940 张图像,不含个人身份信息,通过向世界各地的人征集图像收集而成。我们分析 GeoDE,以理解以此方式收集的图像与网络抓取相比的差异。我们展示了其作为评估和训练数据集的用途,使我们能够突出并开始缓解当前模型的缺陷,尽管 GeoDE 的规模相对较小。我们在 https://geodiverse-data-collection.cs.princeton.edu 发布了完整的数据集和代码
英文摘要:
Current dataset collection methods typically scrape large amounts of data from the web. While this technique is extremely scalable, data collected in this way tends to reinforce stereotypical biases, can contain personally identifiable information, and typically originates from Europe and North America. In this work, we rethink the dataset collection paradigm and introduce GeoDE, a geographically diverse dataset with 61,940 images from 40 classes and 6 world regions, with no personally identifiable information, collected by soliciting images from people around the world. We analyse GeoDE to understand differences in images collected in this manner compared to web-scraping. We demonstrate its use as both an evaluation and training dataset, allowing us to highlight and begin to mitigate the shortcomings in current models, despite GeoDE's relatively small size. We release the full dataset and code at https://geodiverse-data-collection.cs.princeton.edu
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。