论文标题
面部识别：太偏见了，还是不偏见？
Face Recognition: Too Bias, or Not Too Bias?
论文作者
论文摘要
我们揭示了对最先进的面部识别（FR）系统中偏见问题的关键见解，该系统使用野生（BFW）数据集中的新型面孔：对于性别和种族群体平衡的数据。我们显示了跨不同子组的面对对对的最佳得分阈值的变化。因此，学习所有对的全局阈值的常规方法，导致子组之间的性能差距。通过学习特定于亚组的阈值，我们不仅减轻了性能差距中的问题，而且还表现出显着的增强性能。此外，我们进行人类评估以衡量人类的偏见，这支持了这样一种假设，即这种偏见存在于人类的感知中。对于BFW数据库，源代码等等，请访问github.com/visionjo/facerec-bias-bfw。
We reveal critical insights into problems of bias in state-of-the-art facial recognition (FR) systems using a novel Balanced Faces In the Wild (BFW) dataset: data balanced for gender and ethnic groups. We show variations in the optimal scoring threshold for face-pairs across different subgroups. Thus, the conventional approach of learning a global threshold for all pairs resulting in performance gaps among subgroups. By learning subgroup-specific thresholds, we not only mitigate problems in performance gaps but also show a notable boost in the overall performance. Furthermore, we do a human evaluation to measure the bias in humans, which supports the hypothesis that such a bias exists in human perception. For the BFW database, source code, and more, visit github.com/visionjo/facerec-bias-bfw.
