Spatial Attention Supervision for Defect Localization: Exploiting Ground-Truth Masks as Training Signal in Diffusion-Augmented Defect Detection
The results show that existing evaluation masks can act as practical training signals that measurably and reproducibly improve where defect classifiers attend, and contextualizing the supervised gains.
Sajjad Rezvani Boroujeni, Muskan Saraf, Gnana Tulasi Makineni et al.
· 0 citations