Dynamic Snake Upsampling Operater and Boundary-Skeleton Weighted Loss for Tubular Structure Segmentation
This work addresses a specific challenge in medical imaging and other fields where precise segmentation of thin, winding structures is crucial for downstream analysis.
The paper tackles the problem of accurately segmenting slender tubular structures like blood vessels by introducing a dynamic snake upsampling operator that adapts to curvature and a boundary-skeleton weighted loss that balances focus on main bodies and boundaries. Experiments across multiple datasets and networks show improvements in both pixel-wise segmentation accuracy and topological consistency.
Accurate segmentation of tubular topological structures (e.g., fissures and vasculature) is critical in various fields to guarantee dependable downstream quantitative analysis and modeling. However, in dense prediction tasks such as semantic segmentation and super-resolution, conventional upsampling operators cannot accommodate the slenderness of tubular structures and the curvature of morphology. This paper introduces a dynamic snake upsampling operators and a boundary-skeleton weighted loss tailored for topological tubular structures. Specifically, we design a snake upsampling operators based on an adaptive sampling domain, which dynamically adjusts the sampling stride according to the feature map and selects a set of subpixel sampling points along the serpentine path, enabling more accurate subpixel-level feature recovery for tubular structures. Meanwhile, we propose a skeleton-to-boundary increasing weighted loss that trades off main body and boundary weight allocation based on mask class ratio and distance field, preserving main body overlap while enhancing focus on target topological continuity and boundary alignment precision. Experiments across various domain datasets and backbone networks show that this plug-and-play dynamic snake upsampling operator and boundary-skeleton weighted loss boost both pixel-wise segmentation accuracy and topological consistency of results.