Beyond Skin Tone: A Multidimensional Measure of Apparent Skin Color
This addresses fairness issues in computer vision by providing a more comprehensive measure for bias assessment, though it is incremental by building on existing skin tone scales.
The paper tackles the problem of measuring apparent skin color in computer vision beyond just skin tone, introducing a hue angle dimension from red to yellow, which reveals additional biases in datasets and models.
This paper strives to measure apparent skin color in computer vision, beyond a unidimensional scale on skin tone. In their seminal paper Gender Shades, Buolamwini and Gebru have shown how gender classification systems can be biased against women with darker skin tones. Subsequently, fairness researchers and practitioners have adopted the Fitzpatrick skin type classification as a common measure to assess skin color bias in computer vision systems. While effective, the Fitzpatrick scale only focuses on the skin tone ranging from light to dark. Towards a more comprehensive measure of skin color, we introduce the hue angle ranging from red to yellow. When applied to images, the hue dimension reveals additional biases related to skin color in both computer vision datasets and models. We then recommend multidimensional skin color scales, relying on both skin tone and hue, for fairness assessments.