SCRWKV: Ultra-Compact Structure-Calibrated Vision-RWKV for Topological Crack Segmentation
SCRWKV: Ultra-Compact Structure-Calibrated Vision-RWKV for Topological Crack Segmentation
AI总结：
本文提出了一种高效的SCRWKV网络，通过新颖的结构场编码器和轻量级解码器，实现结构裂缝拓扑分割的高精度建模，其在多个复杂纹理和严重干扰的基准测试中表现出色，参数量仅为1.22M，达到了84.28%的F1分数和85.12%的mIoU。
摘要：
实现跨多样场景的结构裂缝像素级准确分割仍然是一个严峻的挑战。现有方法在平衡裂缝拓扑建模与计算效率之间面临显著瓶颈，往往无法在高分割质量与低资源需求之间取得平衡。为了解决这些限制，我们提出了Ultra-Compact Structure-Calibrated Vision RWKV (SCRWKV)，一种通过新颖的Structure-Field Encoder (SFE) backbone实现高精度建模的网络，同时保持线性复杂度。SFE集成了Adaptive Multi-scale Cascaded Modulator (AMCM)以增强纹理表示，并利用Structure-Calibrated Insight Unit (SCIU)作为其核心引擎。具体而言，SCIU采用Geometry-guided Bidirectional Structure Transformation (GBST)来捕捉拓扑相关性，并将Dynamic Self-Calibrating Decay (DSCD)整合到Dy-WKV中以抑制噪声传播。此外，我们引入了一种轻量级的Cross-Scale Harmonic Fusion (CSHF)解码器以实现精确的特征聚合。系统评估表明，在多个具有复杂纹理和严重干扰的基准测试中，仅拥有1.22M参数的SCRWKV显著优于现有最佳方法。在TUT数据集上，该模型达到了84.28%的F1分数和85.12%的mIoU，证明了其在高效现实部署中的鲁棒潜力。代码可在https://github.com/zhxhzy/SCRWKV上获取。
英文摘要：
Achieving pixel-level accurate segmentation of structural cracks across diverse scenarios remains a formidable challenge. Existing methods face significant bottlenecks in balancing crack topology modeling with computational efficiency, often failing to reconcile high segmentation quality with low resource demands. To address these limitations, we propose the Ultra-Compact Structure-Calibrated Vision RWKV (SCRWKV), a network that achieves high-precision modeling via a novel Structure-Field Encoder (SFE) backbone while maintaining linear complexity. The SFE integrates the Adaptive Multi-scale Cascaded Modulator (AMCM) to enhance texture representation and utilizes the Structure-Calibrated Insight Unit (SCIU) as its core engine. Specifically, the SCIU employs the Geometry-guided Bidirectional Structure Transformation (GBST) to capture topological correlations and integrates the Dynamic Self-Calibrating Decay (DSCD) into Dy-WKV to suppress noise propagation. Furthermore, we introduce a lightweight Cross-Scale Harmonic Fusion (CSHF) decoder to achieve precise feature aggregation. Systematic evaluations on multiple benchmarks characterized by complex textures and severe interference demonstrate that SCRWKV, with only 1.22M parameters, significantly outperforms SOTA methods. Achieving an F1 score of 0.8428 and mIoU of 0.8512 on the TUT dataset, the model confirms its robust potential for efficient real-world deployment. The code is available at https://github.com/zhxhzy/SCRWKV.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
