<p>Organisations must share facial imagery that remains useful for analysis while protecting identity. Many current methods fail to strike this balance: reconstruction-centred encoder–decoder designs tend to blur salient detail, whereas latent edits in pretrained generators often retain or drift identity cues, undermining privacy and utility. We present ReFaceX, a reversible anonymisation framework that separates what to protect from what to preserve. A donor identity code steers a U-Net anonymiser with Identity Feature Fusion to change identity while retaining non-identity content such as pose, background and expression. A learned steganographic channel carries a compact recovery payload, and reconstruction gradients are blocked at the stego image so the anonymiser is never rewarded for keeping identity. The threat model is stated explicitly and outcomes are audited with strong recognisers. On LFW and CelebA-HQ datasets at <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(256\times 256\)</EquationSource> </InlineEquation>, ReFaceX reduces identity similarity across FaceNet, ArcFace and AdaFace, and improves recovered-image quality (SSIM <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(0.9378\)</EquationSource> </InlineEquation>, LPIPS <InlineEquation ID="IEq3"> <EquationSource Format="TEX">\(0.1002\)</EquationSource> </InlineEquation>, PSNR <InlineEquation ID="IEq4"> <EquationSource Format="TEX">\(23.97\)</EquationSource> </InlineEquation> dB), while operating in real time on a single RTX&#xa0;3090. Robustness to common JPEG re-encoding is also demonstrated. By turning the privacy–utility balance into an explicit and auditable operating choice, ReFaceX provides a practical template for responsible release of facial imagery and a foundation for extensions to video, higher resolutions and stronger recovery guarantees.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

ReFaceX: donor-driven reversible face anonymisation with detached recovery

  • Dost Muhammad,
  • Muhammad Salman,
  • Syed Muhammad Haider Shah,
  • Malika Bendechache

摘要

Organisations must share facial imagery that remains useful for analysis while protecting identity. Many current methods fail to strike this balance: reconstruction-centred encoder–decoder designs tend to blur salient detail, whereas latent edits in pretrained generators often retain or drift identity cues, undermining privacy and utility. We present ReFaceX, a reversible anonymisation framework that separates what to protect from what to preserve. A donor identity code steers a U-Net anonymiser with Identity Feature Fusion to change identity while retaining non-identity content such as pose, background and expression. A learned steganographic channel carries a compact recovery payload, and reconstruction gradients are blocked at the stego image so the anonymiser is never rewarded for keeping identity. The threat model is stated explicitly and outcomes are audited with strong recognisers. On LFW and CelebA-HQ datasets at \(256\times 256\) , ReFaceX reduces identity similarity across FaceNet, ArcFace and AdaFace, and improves recovered-image quality (SSIM \(0.9378\) , LPIPS \(0.1002\) , PSNR \(23.97\) dB), while operating in real time on a single RTX 3090. Robustness to common JPEG re-encoding is also demonstrated. By turning the privacy–utility balance into an explicit and auditable operating choice, ReFaceX provides a practical template for responsible release of facial imagery and a foundation for extensions to video, higher resolutions and stronger recovery guarantees.