发现大图分割中伪影源于归一化层，提出有效解决方案。
Tiling artifacts and trade-offs of feature normalization in the segmentation of large biological images
- 分析归一化层是导致大图分割伪影的根源
- 验证BatchRenorm能消除伪影并提升跨数据集性能
- 适合从事生物图像分割的研究者参考
大型图像分割在显微成像、医学影像和遥感中常见。通常采用滑动窗口推理，理论上可无缝拼接预测结果，但现有许多主流流程仍存在拼接伪影。我们探究其根本原因，发现源自神经网络中的归一化层。提出检测归一化问题的指标，并以三个不同显微成像数据集为例，研究无伪影与高质量预测之间的权衡。最终建议使用BatchRenorm作为最优归一化策略，能有效消除拼接伪影，提升模型迁移能力，增强训练网络在新数据集上的可复用性。
原文摘要 · Abstract (English)
Segmentation of very large images is a common problem in microscopy, medical imaging or remote sensing. The problem is usually addressed by sliding window inference, which can theoretically lead to seamlessly stitched predictions. However, in practice many of the popular pipelines still suffer from tiling artifacts. We investigate the root cause of these issues and show that they stem from the normalization layers within the neural networks. We propose indicators to detect normalization issues and further explore the trade-offs between artifact-free and high-quality predictions, using three diverse microscopy datasets as examples. Finally, we propose to use BatchRenorm as the most suitable normalization strategy, which effectively removes tiling artifacts and enhances transfer performance, thereby improving the reusability of trained networks for new datasets.
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。