用少量数据生成逼真纳米颗粒图像，提升分割模型训练效果
F-ANcGAN: An Attention-Enhanced Cycle Consistent Generative Adversarial Architecture for Synthetic Image Generation of Nanoparticles
- 引入注意力增强的循环一致性生成框架，从分割图生成SEM图像
- 对二氧化钛数据集生成的原始FID为17.65，后处理后降至10.39
- 适合纳米材料分析、数据稀缺场景下的合成数据生成
纳米材料研究在能源、医学和材料科学中日益重要，准确分析纳米颗粒拓扑结构对其性能至关重要。然而，高质量标注数据集的缺乏严重制约了纳米尺度成像分割模型的构建。为此，我们提出F-ANcGAN，一种注意力增强的循环一致性生成对抗网络，可在少量样本下训练，直接从分割图生成真实的扫描电子显微镜（SEM）图像。模型采用风格U-Net生成器与带自注意力的U-Net分割网络，以捕捉结构关联，并结合数据增强提升数据多样性。在二氧化钛（TiO₂）数据集上，生成图像的原始FID得分为17.65，经高效后处理后降至接近10.39。该方法可规模化生成高保真合成数据集，显著提升下游分割任务的训练效果，有效缓解纳米颗粒分析中的数据短缺问题，拓展至资源受限领域应用。
原文摘要 · Abstract (English)
Nanomaterial research is becoming a vital area for energy, medicine, and materials science, and accurate analysis of the nanoparticle topology is essential to determine their properties. Unfortunately, the lack of high-quality annotated datasets drastically hinders the creation of strong segmentation models for nanoscale imaging. To alleviate this problem, we introduce F-ANcGAN, an attention-enhanced cycle consistent generative adversarial system that can be trained using a limited number of data samples and generates realistic scanning electron microscopy (SEM) images directly from segmentation maps. Our model uses a Style U-Net generator and a U-Net segmentation network equipped with self-attention to capture structural relationships and applies augmentation methods to increase the variety of the dataset. The architecture reached a raw FID score of 17.65 for TiO$_2$ dataset generation, with a further reduction in FID score to nearly 10.39 by using efficient post-processing techniques. By facilitating scalable high-fidelity synthetic dataset generation, our approach can improve the effectiveness of downstream segmentation task training, overcoming severe data shortage issues in nanoparticle analysis, thus extending its applications to resource-limited fields.
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。