用截断扩散模型加速盲人脸修复，速度提升4.75倍
TD-BFR: Truncated Diffusion Model for Efficient Blind Face Restoration
- 分三阶段从低清到高清逐步修复，结合截断采样提速
- 相比现有方法平均快4.75倍，修复质量仍保持领先
- 适合需要快速修复人脸的实时应用或资源受限场景
基于扩散模型的人脸盲修复展现了强大生成能力，但普遍存在训练与推理速度慢、细节恢复不足的问题。为此，我们提出一种面向高效盲人脸修复的截断扩散模型（TD-BFR），采用分阶段渐进式修复策略。该方法通过创新的截断采样机制，从低分辨率低质量图像开始加速采样，并引入自适应退化消除模块以处理未知退化，连接多分辨率生成流程。同时，对预训练扩散模型先验进行适配，以恢复丰富面部细节。实验表明，TD-BFR在平均速度上比当前最优扩散类方法快4.75倍，同时保持优异的修复质量。
原文摘要 · Abstract (English)
Diffusion-based methodologies have shown significant potential in blind face restoration (BFR), leveraging their robust generative capabilities. However, they are often criticized for two significant problems: 1) slow training and inference speed, and 2) inadequate recovery of fine-grained facial details. To address these problems, we propose a novel Truncated Diffusion model for efficient Blind Face Restoration (TD-BFR), a three-stage paradigm tailored for the progressive resolution of degraded images. Specifically, TD-BFR utilizes an innovative truncated sampling method, starting from low-quality (LQ) images at low resolution to enhance sampling speed, and then introduces an adaptive degradation removal module to handle unknown degradations and connect the generation processes across different resolutions. Additionally, we further adapt the priors of pre-trained diffusion models to recover rich facial details. Our method efficiently restores high-quality images in a coarse-to-fine manner and experimental results demonstrate that TD-BFR is, on average, \textbf{4.75$\times$} faster than current state-of-the-art diffusion-based BFR methods while maintaining competitive quality.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。