MixerCSeg: An Efficient Mixer Architecture for Crack Segmentation via Decoupled Mamba Attention

TL;DR

MixerCSeg introduces a novel, efficient mixer architecture combining CNN, Transformer, and Mamba-inspired pathways to improve pixel-level crack segmentation by capturing local textures, global dependencies, and sequential context.

Contribution

The paper proposes MixerCSeg, a new encoder architecture that integrates multiple pathways and a TransMixer core to enhance crack segmentation accuracy and efficiency.

Findings

Achieves state-of-the-art performance on crack segmentation benchmarks.

Operates with only 2.05 GFLOPs and 2.54 million parameters.

Demonstrates strong generalization and structural fidelity.

Abstract

Feature encoders play a key role in pixel-level crack segmentation by shaping the representation of fine textures and thin structures. Existing CNN-, Transformer-, and Mamba-based models each capture only part of the required spatial or structural information, leaving clear gaps in modeling complex crack patterns. To address this, we present MixerCSeg, a mixer architecture designed like a coordinated team of specialists, where CNN-like pathways focus on local textures, Transformer-style paths capture global dependencies, and Mamba-inspired flows model sequential context within a single encoder. At the core of MixerCSeg is the TransMixer, which explores Mamba's latent attention behavior while establishing dedicated pathways that naturally express both locality and global awareness. To further enhance structural fidelity, we introduce a spatial block processing strategy and a Direction-guided Edge Gated Convolution (DEGConv) that strengthens edge sensitivity under irregular crack geometries with minimal computational overhead. A Spatial Refinement Multi-Level Fusion (SRF) module is then employed to refine multi-scale details without increasing complexity. Extensive experiments on multiple crack segmentation benchmarks show that MixerCSeg achieves state-of-the-art performance with only 2.05 GFLOPs and 2.54 M parameters, demonstrating both efficiency and strong representational capability. The code is available at https://github.com/spiderforest/MixerCSeg.