Training-Free Tunnel Defect Inspection and Engineering Interpretation via Visual Recalibration and Entity Reconstruction
For tunnel inspection engineers, this work provides a training-free method to produce structured defect evidence (location, severity, etc.) from foundation models, addressing the gap between coarse proposals and practical engineering documentation.
TunnelMIND is a training-free framework for tunnel defect inspection that recalibrates coarse language-guided proposals via dense visual consistency and reconstructs them into structured defect entities with attributes like category, location, and severity. It achieves F1 scores of 0.68, 0.78, and 0.72 on visible, GPR, and road defect tasks, respectively, enabling engineering-readable reports without training.
Tunnel inspection requires outputs that can support defect localization, measurement, severity grading, and engineering documentation. Existing training-free foundation-model pipelines usually stop at coarse open-vocabulary proposals, which are difficult to use directly in interference-heavy tunnel scenes. We propose a training-free framework TunnelMIND. Specifically, language-guided defect proposals are not treated as final outputs; instead, their spatial support is recalibrated at inference time through dense visual consistency, so that coarse semantic anchors can be transformed into more reliable prompts under tunnel-specific hard negatives. The resulting masks are further reconstructed into structured defect entities with category, location, geometry, severity, and context attributes, which are then mapped to retrieval-grounded explanation and engineering-readable report generation under expert knowledge constraints. On visible, GPR, and road defect tasks, TunnelMIND achieves F1 scores of 0.68, 0.78, and 0.72, respectively. Overall, TunnelMIND shows that training-free tunnel inspection can move beyond coarse localization toward structured defect evidence for engineering assessment.