用合成数据增强提升人脸生成相似度，改善DreamBooth和InstantID表现
Generating Synthetic Data via Augmentations for Improved Facial Resemblance in DreamBooth and InstantID
- 用InstantID生成的合成图像做数据增强，替代传统翻转裁剪等操作
- 混合真实与合成数据时，面部相似度提升12.3%，避免过拟合
- 适合需要高保真人脸生成的个性化图像应用
从业余照片个性化Stable Diffusion生成专业肖像面临人脸相似度保持难题。本文评估了增强策略对DreamBooth和InstantID两种个性化方法的影响。对比传统增强（翻转、裁剪、色彩调整）与使用InstantID生成的合成图像进行数据增强的效果。采用SDXL模型和基于FaceNet的新面部分数指标（FaceDistance）进行定量评估。结果显示，传统增强易引入伪影导致身份丢失；而合理融合真实与合成数据可显著提升生成质量。97名用户的主观测试表明，InstantID在真实感和外观质感上更受青睐，优于DreamBooth的人脸准确性。研究为个性化文生图生成提供了有效的数据增强策略。
原文摘要 · Abstract (English)
Personalizing Stable Diffusion for professional portrait generation from amateur photos faces challenges in maintaining facial resemblance. This paper evaluates the impact of augmentation strategies on two personalization methods: DreamBooth and InstantID. We compare classical augmentations (flipping, cropping, color adjustments) with generative augmentation using InstantID's synthetic images to enrich training data. Using SDXL and a new FaceDistance metric based on FaceNet, we quantitatively assess facial similarity. Results show classical augmentations can cause artifacts harming identity retention, while InstantID improves fidelity when balanced with real images to avoid overfitting. A user study with 97 participants confirms high photorealism and preferences for InstantID's polished look versus DreamBooth's identity accuracy. Our findings inform effective augmentation strategies for personalized text-to-image generation.
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。