titleSynthFaces: Balanced Large Scale Human Dataset
authorsSergey Karpukhin, Denis Bobkov, Kirill Morozov, Aibek Alanov
date2026-10-29
venuearXiv (preprint, not peer reviewed)
idarXiv:2610.40119
upvotes1 (Hugging Face Daily Papers, 2026-11-08)
linksabs pdf hf code
cite
abstractThe pursuit of fair and equitable AI in human-centric tasks is critically hindered by the inherent biases and privacy risks of real-world datasets, which offer imbalanced representations of identity and often contain sensitive personal information. Consequently, state-of-the-art generative models trained on this skewed data exhibit unfair performance, particularly in identity-preserving generation. In this work we propose a pipeline for controllable generation of dataset with diverse faces for generative models training and finetuning. We achieve this by expanding foundational Arc2Face diffusion model with a flow matching generative model in the manifold of identity embeddings. We demonstrate that such approach achieves both goals of generative fidelity and label diversity not present in real human-centric datasets. As a result we collect and further contextualize a highly balanced identity-centric dataset of humans named SynthFaces. By design, our released images and identities are synthetic, reducing direct reliance on real personal images; we empirically check identity separation with a nearest-neighbor non-leakage metric. We purpose it for training and tuning of fair models for various downstream tasks, and showcase its usability in finetuning an existing adapter-based method for personalized generation.

bibtex

@misc{karpukhin2026synthfaces,
  title={SynthFaces: Balanced Large Scale Human Dataset},
  author={Sergey Karpukhin and Denis Bobkov and Kirill Morozov and Aibek Alanov},
  year={2026},
  eprint={2610.40119},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2610.40119}
}