arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.11097cs.CVcs.LGstat.ML
收藏

目标检测中的预测不等式

Predictive Inequity in Object Detection

Georgia Tech(佐治亚理工学院)

Benjamin Wilson, Judy Hoffman, Jamie Morgenstern

摘要:

在这项工作中,我们调查国家的最先进的目标检测系统是否有公平的预测性能不同肤色的行人。这项工作的动机是最近的许多ML和视觉系统的例子显示出较高的出错率的某些人口群体比其他人。我们注释现有的大规模数据集,其中包含行人,BDD 100 K,Fitzpatrick肤色范围[1-3]或[4-6]。然后,我们对这两种肤色分组之间的性能进行了深入的比较分析,发现一天中的时间和遮挡都不能解释这种行为,这表明这种差异不仅仅是4-6范围内的行人出现在更难检测的场景中的结果。我们调查了一天中的时间,遮挡和重新加权训练过程中的监督损失在多大程度上影响了这种预测偏差。

英文摘要:

In this work, we investigate whether state-of-the-art object detection systems have equitable predictive performance on pedestrians with different skin tones. This work is motivated by many recent examples of ML and vision systems displaying higher error rates for certain demographic groups than others. We annotate an existing large scale dataset which contains pedestrians, BDD100K, with Fitzpatrick skin tones in ranges [1-3] or [4-6]. We then provide an in-depth comparative analysis of performance between these two skin tone groupings, finding that neither time of day nor occlusion explain this behavior, suggesting this disparity is not merely the result of pedestrians in the 4-6 range appearing in more difficult scenes for detection. We investigate to what extent time of day, occlusion, and reweighting the supervised loss during training affect this predictive bias.

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。

↑