Investigating Bias in Deep Face Analysis: The KANFace Dataset and Empirical Study

Deep learning-based methods have pushed the limits of the state-of-the-art in\nface analysis. However, despite their success, these models have raised\nconcerns regarding their bias towards certain demographics. This bias is\ninflicted both by limited diversity across demographics in the training set, as\nwell as the design of the algorithms. In this work, we investigate the\ndemographic bias of deep learning models in face recognition, age estimation,\ngender recognition and kinship verification. To this end, we introduce the most\ncomprehensive, large-scale dataset of facial images and videos to date. It\nconsists of 40K still images and 44K sequences (14.5M video frames in total)\ncaptured in unconstrained, real-world conditions from 1,045 subjects. The data\nare manually annotated in terms of identity, exact age, gender and kinship. The\nperformance of state-of-the-art models is scrutinized and demographic bias is\nexposed by conducting a series of experiments. Lastly, a method to debias\nnetwork embeddings is introduced and tested on the proposed benchmarks.\n