Abstract
Blind face restoration (BFR) aims to recover high-quality facial images from degraded inputs, yet its inherently ill-posed nature leads to ambiguous and uncontrollable solutions. Recent diffusion-based BFR methods improve perceptual quality but remain uncontrollable, whereas text-guided face editing enables attribute manipulation without reliable restoration. To address these issues, we propose A$^2$BFR, an attribute-aware blind face restoration framework that unifies high-fidelity reconstruction with prompt-controllable generation. Built upon a Diffusion Transformer backbone with unified image-text cross-modal attention, A$^2$BFR jointly conditions the denoising trajectory on both degraded inputs and textual prompts. To inject semantic priors, we introduce attribute-aware learning, which supervises denoising latents using facial attribute embeddings extracted by an attribute-aware encoder. To further enhance prompt controllability, we introduce semantic dual-training, which leverages the pairwise attribute variations in our newly curated AttrFace-90K dataset to enforce attribute discrimination while preserving fidelity. Extensive experiments demonstrate that A$^2$BFR achieves state-of-the-art performance in both restoration fidelity and instruction adherence, outperforming diffusion-based BFR baselines by -0.0467 LPIPS and +52.58% attribute accuracy, while enabling fine-grained, prompt-controllable restoration even under severe degradations.
Abstract (translated)
盲人脸修复（BFR）旨在从退化输入中恢复高质量人脸图像，但其固有的病态性会导致模糊且不可控的结果。近期基于扩散模型的BFR方法提升了感知质量，但仍缺乏可控性；而文本引导的人脸编辑虽能实现属性操控，却难以保证可靠的修复效果。针对这些问题，我们提出了A²BFR——一种属性感知的盲人脸修复框架，实现了高保真重建与提示可控生成的统一。该框架基于扩散Transformer主干网络，通过统一的图像-文本跨模态注意力机制，使去噪轨迹同时依赖于退化输入与文本提示。为注入语义先验，我们引入属性感知学习，利用属性感知编码器提取的人脸属性嵌入来监督去噪隐空间。为进一步提升提示可控性，我们提出语义双重训练策略，利用新构建的AttrFace-90K数据集中的成对属性变化，在保持保真度的同时强化属性判别能力。大量实验表明，A²BFR在修复保真度与指令遵循性上均达到最优性能，相比基于扩散的BFR基线方法，LPIPS指标降低0.0467，属性准确率提升52.58%，并能在严重退化条件下实现细粒度、提示可控的修复。
URL
https://arxiv.org/abs/2603.29423
