SOURCE-LINKED INTELLIGENCE

CoRe-SAM3: Conditional Semantic--Visual Reconciliation for SAM3 Crack Segmentation

arXiv · AI, language, vision and robotics · article · Sep 5, 2026 · UTC

Crack segmentation requires a model to recognize target semantics while accurately recovering thin, low-contrast, and topologically continuous local structures. Although SAM3 provides strong open-concept segmentation, its direct application to the crack domain still misses weak cracks, activates crack-like background regions, and produces local boundary errors. We first diagnose the functional differences between the internal prompt-conditioned semantic representation and native visual representation of SAM3 on five crack datasets. The results show that the semantic representation already carr

Read original source ↗ Open in workspace

recordType

paper

region

Global