基于希尔伯特曲线的注意力实现语义分割网络的拓扑保持图像张量表示
Hilbert Curve-Based Attention Enabling Topology-Preserving Image Tensor Representation for Semantic Segmentation Network
把 Hilbert 曲线空间填充顺序引入解码器的注意力计算，用一维连续索引保持二维张量的空间邻接关系，可作为通用手段替换分割/检测解码器中的展平或窗口划分方式，以改善边界与细结构的连续性。
原文依据 Its decoder incorporates a Hilbert curve-based topology-preserving mechanism to maintain spatial continuity and boundary precision
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论由于表面纹理复杂且光照变化显著，基于无人机的建筑缺陷分割仍具挑战性。我们提出 TPSegformer，一种拓扑保持的分割框架，可缓解此类场景中的误分割问题。其解码器引入基于希尔伯特曲线的拓扑保持机制，在类别层计算过程中维持空间连续性与边界精度。轻量级多尺度融合模块增强语义表示，全局上下文建模则强化整体感知。在建筑缺陷数据集上的实验表明，TPSegformer 优于现有分割方法，达到 80.77% 平均交并比和 90.22% 准确率。在 Dacl10k 数据集上，它保持了较强的泛化能力，在多种材料与缺陷类型上达到 44.27% 平均交并比和 60.32% 准确率。代码见：https://github.com/mumu-k/TPSegformer
英文原文摘要 894 字符
Drone-based building defect segmentation remains challenging due to complex surface textures and illumination variations. We propose TPSegformer, a topology-preserving segmentation framework that mitigates mis-segmentation in such scenarios. Its decoder incorporates a Hilbert curve-based topology-preserving mechanism to maintain spatial continuity and boundary precision during category layer computation. A lightweight multi-scale fusion module enhances semantic representation, while global context modeling strengthens holistic perception. Experiments on the building defect dataset show that TPSegformer outperforms existing segmentation methods, achieving 80.77% mIoU and 90.22% Acc. On the Dacl10k dataset, it maintains strong generalization, reaching 44.27% mIoU and 60.32% Acc across diverse materials and defect types. The code is available at : https://github.com/mumu-k/TPSegformer