Training-Free Tunnel Defect Inspection and Engineering Interpretation via Visual Recalibration and Entity Reconstruction

arXiv:2604.27928v1 Announce Type: cross Abstract: 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, the