arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.02475cs.CVcs.LG
收藏

大规模数据集与基准用于野外皮肤色调分类

Skin Tone in the Wild: A Dataset and Benchmark for 10-tone MST Classification

Vitor Matias, Márcus Vinícius Lobo Costa, Tiago Novello, João Batista Neto

AI总结:

本文提出STW大规模数据集和SkinToneNet模型,通过基准测试展示了深度学习在皮肤色调分类中的有效性,并实现了对公开数据集的公平性评估

摘要:

深度学习模型往往继承自训练数据中的偏见。尽管性别和种族的公平性已被广泛研究,但细粒度的皮肤色调分析仍然具有挑战性,因为缺乏粒度精细的标注数据集。现有方法通常依赖于医学6级Fitzpatrick皮肤色级,缺乏视觉代表性,或使用小规模的私人数据集,这阻碍了可重复性,或经常依赖于经典计算机视觉流程,只有少数使用深度学习。它们忽略了训练-测试泄漏和数据集不平衡等问题,并受到小规模或不可用数据集的限制。在本工作中,我们提出了一个全面的皮肤色调公平性框架。首先,我们介绍了STW,一个大规模、开放获取的数据集,包含来自3564人的42313张图像,使用10级MST皮肤色级进行标注。其次,我们对经典计算机视觉(SkinToneCCV)和深度学习方法进行了基准测试,证明经典模型提供接近随机的结果,而深度学习方法达到接近标注者准确度。最后,我们提出了SkinToneNet,一个经过微调的ViT模型,能够在领域外数据上实现最先进的泛化能力,这使公共数据集如CelebA和VGGFace2的公平性审计变得可靠。本工作在皮肤色调分类和公平性评估方面提供了最先进的结果。代码和数据即将发布

英文摘要:

Skin tone classification is a fundamental task in computer vision, with applications in quantifying medical features, assessing demographic representativeness, and mitigating algorithmic bias. However, existing datasets either rely on a restrictive 6-tone Fitzpatrick scale, prohibit training through restrictive licenses, or lack representation of extreme skin tones, while current methods often fail to disentangle true skin tone from in-the-wild imaging conditions, degrading to random-classifier performance. We introduce skin tone in the wild (STW), the first open-access dataset annotated with the 10-tone Monk Skin Tone (MST) scale, containing over 42K images and specifically curated to include underrepresented tones. Alongside a rigorous evaluation protocol, we propose SkinToneNet, a ViT-based model that leverages global image context for robust in-the-wild generalization. Experiments show that SkinToneNet achieves 90% weighted off-by-one accuracy, substantially outperforming prior baselines. We further use SkinToneNet to audit widely used face datasets, including CelebA and VGGFace2, revealing substantial representation imbalances across MST tones.

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。

↑