Tiling artifacts and trade-offs of feature normalization in the segmentation of large biological images
This addresses a practical problem for researchers in microscopy, medical imaging, and remote sensing, offering an incremental improvement to existing segmentation pipelines.
The paper tackled tiling artifacts in large image segmentation caused by normalization layers, proposing BatchRenorm as a solution that removes artifacts and improves transfer performance across three microscopy datasets.
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.