用扩散模型提升模糊噪声等劣质人脸图像的识别率
Latent Diffusion Based Face Enhancement under Degraded Conditions for Forensic Face Recognition
- 基于潜空间扩散模型，结合面部放大LoRA微调增强低质人脸
- 识别准确率从29.1%提升至84.5%，改善55.4个百分点
- 适用于法医人脸识别场景，尤其在压缩、模糊、噪声下表现突出
当处理低质量法医取证图像时，人脸识别系统性能严重下降。本文评估了基于潜空间扩散模型的人脸增强方法在法医相关退化条件下的有效性。基于包含3000名个体、共24000次识别尝试的LFW数据集，采用Flux.1 Kontext Dev流程并结合Facezoom LoRA适配，测试了七类退化情况，包括压缩伪影、模糊效应和噪声污染。实验结果表明，该方法显著提升识别准确率，整体识别率从29.1%提高至84.5%（提升55.4个百分点，95%置信区间：[54.1, 56.7]）。统计分析显示，所有退化类型下性能均有显著提升，效应量超过实际意义的常规阈值。研究证实了先进扩散模型在法医人脸识别中的应用潜力。
原文摘要 · Abstract (English)
Face recognition systems experience severe performance degradation when processing low-quality forensic evidence imagery. This paper presents an evaluation of latent diffusion-based enhancement for improving face recognition under forensically relevant degradations. Using a dataset of 3,000 individuals from LFW with 24,000 recognition attempts, we implement the Flux.1 Kontext Dev pipeline with Facezoom LoRA adaptation to test against seven degradation categories, including compression artefacts, blur effects, and noise contamination. Our approach demonstrates substantial improvements, increasing overall recognition accuracy from 29.1% to 84.5% (55.4 percentage point improvement, 95% CI: [54.1, 56.7]). Statistical analysis reveals significant performance gains across all degradation types, with effect sizes exceeding conventional thresholds for practical significance. These findings establish the potential of sophisticated diffusion based enhancement in forensic face recognition applications.
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。