Research map / Detection and segmentation / CNN features
MixerCSeg: An Efficient Mixer Architecture for Crack Segmentation via Decoupled Mamba Attention
Zilong Zhao, Zhengming Ding, Pei Niu, Wenhao Sun, Feng Guo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add TransMixer to the VMamba + Segformer baseline (ablation) | baseline crack segmenter · DeepCrack | mIoU: 0.8826 -> 0.9016 | Improved |
| add DEGConv on top of TransMixer (ablation) | MixerCSeg · DeepCrack | mIoU: 0.9016 -> 0.9097 | Improved |
| add SRF module replacing the Segformer decoder (ablation) | MixerCSeg · DeepCrack | mIoU: 0.9097 -> 0.9151 | Improved |
| add SRF module replacing the Segformer decoder (ablation) | MixerCSeg · computational cost | FLOPs (G): 19.18 -> 2.05 | Improved |
| use TransMixer instead of MambaVision hybrid encoder (ablation) | MixerCSeg · DeepCrack | mIoU: 0.9020 -> 0.9151 | Improved |
| use TransMixer instead of RestorMixer hybrid encoder (ablation) | MixerCSeg · CamCrack789 | OIS: 0.8206 -> 0.8202 | No clear effect |
| set global channel ratio gamma to 0.5 instead of 0.3, 0.7 or 0.9 (ablation) | TransMixer · DeepCrack | mIoU: 0.9098 (0.3) -> 0.9151 (0.5) | Improved |
| use 8x8 DEGConv cell instead of 2x2 (ablation) | DEGConv · DeepCrack | mIoU: 0.9104 -> 0.9151 | Improved |
| use 180 direction bins instead of 9 (ablation) | DEGConv · DeepCrack | mIoU: 0.9067 -> 0.9151 | Improved |
| use 180 bins instead of 360 bins (ablation) | DEGConv · DeepCrack | mIoU: 0.9077 -> 0.9151 | Improved |
| build direction embedding from directional priors instead of ARConv predicted angles (ablation) | DEGConv · DeepCrack | mIoU: 0.9106 -> 0.9151 | Improved |
| MixerCSeg versus SCSegamba on Crack500 | MixerCSeg · Crack500 | mIoU: 0.7778 -> 0.7824 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SCSegamba — SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures (CVPR 2025) · better on 10 of 10 shared results
- RINDNet — RINDNet: Edge Detection for Discontinuity in Reflectance, Illumination, Normal and Depth (ICCV 2021) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures (CVPR 2025 · similarity 0.90)
- Mind Marginal Non-Crack Regions: Clustering-Inspired Representation Learning for Crack Segmentation (CVPR 2024 · similarity 0.85)
- Cracks in the Foundation: A Civil Infrastructure Dataset to Challenge Vision Foundation Models (NeurIPS 2026 · similarity 0.83)
- Spatial-Mamba: Effective Visual State Space Models via Structure-Aware State Fusion (ICLR 2025 · similarity 0.81)
- Wavelet and Prototype Augmented Query-based Transformer for Pixel-level Surface Defect Detection (CVPR 2025 · similarity 0.81)
- Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection (AAAI 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
