用少量真实图像生成合成数据，提升罕见视觉类别的识别效果。
Few-Shot Synthetic Data Generation with Diffusion Models for Downstream Vision Tasks
- 用LoRA微调扩散模型，仅需20-50张真实图像生成合成样本。
- 在两种不同领域中，合成数据使罕见类别召回率和F1值显著提升。
- 适度合成比例效果最佳，过量反而收益递减，适合数据稀缺场景。
类别不平衡是视觉识别中的长期挑战，尤其在安全关键领域，正样本收集成本高且稀有事件天然稀缺。本文提出一种轻量级合成数据增强流程：仅用20-50张罕见类别的真实图像微调LoRA适配器，并利用预训练扩散模型生成合成样本用于训练。我们在两个结构不同的领域系统评估该方法：胸部X光病理分类（NIH ChestX-ray14）与工业表面裂纹检测（Magnetic Tile Defect dataset）。所有评估均基于独立测试集的真实图像进行。在两个领域中，相比仅使用真实数据训练，合成数据增强均显著提升了罕见类别的召回率和F1分数。性能随适度的合成数据比例提升，但合成比例过高时收益开始递减。结果表明，经LoRA适配的扩散模型为稀有类别提供了一种简单且可扩展的增强机制，可在异构视觉任务中有效应对数据稀缺问题。
原文摘要 · Abstract (English)
Class imbalance is a persistent challenge in visual recognition, particularly in safety-critical domains where collecting positive examples is expensive and rare events are inherently underrepresented. We propose a lightweight synthetic data augmentation pipeline that fine-tunes a LoRA adapter on as few as 20-50 real images of a rare class and uses a pretrained diffusion model to generate synthetic samples for training. We systematically vary the synthetic-to-real ratio and evaluate the approach across two structurally different domains: chest X-ray pathology classification (NIH ChestX-ray14) and industrial surface crack detection (Magnetic Tile Defect dataset). All evaluations are performed on held-out sets of real images only. Across both domains, synthetic augmentation consistently improves rare-class recall and F1 compared to training with real data alone. Performance improves with moderate synthetic augmentation and shows diminishing returns as the synthetic ratio increases. These results suggest that LoRA-adapted diffusion models provide a simple and scalable mechanism for augmenting rare classes, enabling effective learning in data-scarce scenarios across heterogeneous visual domains.
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。