What question did this study set out to answer?

The goal is to enhance pavement crack semantic segmentation by overcoming limitations of existing models in handling complex backgrounds and variable crack shapes.

July 26, 2026Remote SensingOpen Access

GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation

View Full Paper

Ask AI

Bookmark

Authors

SLSangning Li BLBin Liu HGHaiyan Guan

Discussion

Member takes

Overview

Randomized trial demonstrates improved pavement crack segmentation using GeoMamba in diverse road environments, suggesting robust performance across materials.

Key Points

The goal is to enhance pavement crack semantic segmentation by overcoming limitations of existing models in handling complex backgrounds and variable crack shapes.
Developed a Multi-Scale Deformable Visual State Space module for extracting adaptable contextual features.
Introduced a Geometric-Topology Prior Injection module that infuses structural priors and mitigates artifacts.
Performed experiments on DeepCrack and Concrete3K datasets to validate performance.
GeoMamba achieved an mIoU of 83.79% on the DeepCrack dataset.
F1 score reached 89.27%, indicating superior semantic segmentation performance.
Demonstrated robust generalization across various pavement materials.

Cite This Study

Li et al. (2026) studied this question.

synapsesocial.com/papers/6a65a890d3aea3239cd78e0b https://doi.org/10.3390/rs18152449

View Full Paper

Ask AI

Bookmark

July 26, 2026Remote SensingOpen Access

GeoMamba: Geometric-Prior-Infused Multi-Scale Deformable Visual Mamba for Crack Semantic Segmentation

View Full Paper

Ask AI

Bookmark

Authors

SLSangning Li BLBin Liu HGHaiyan Guan

Discussion

Member takes

Overview

Randomized trial demonstrates improved pavement crack segmentation using GeoMamba in diverse road environments, suggesting robust performance across materials.

Key Points

The goal is to enhance pavement crack semantic segmentation by overcoming limitations of existing models in handling complex backgrounds and variable crack shapes.
Developed a Multi-Scale Deformable Visual State Space module for extracting adaptable contextual features.
Introduced a Geometric-Topology Prior Injection module that infuses structural priors and mitigates artifacts.
Performed experiments on DeepCrack and Concrete3K datasets to validate performance.
GeoMamba achieved an mIoU of 83.79% on the DeepCrack dataset.
F1 score reached 89.27%, indicating superior semantic segmentation performance.
Demonstrated robust generalization across various pavement materials.

Cite This Study

Li et al. (2026) studied this question.

synapsesocial.com/papers/6a65a890d3aea3239cd78e0b https://doi.org/10.3390/rs18152449

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark