industrial inspection 1multi-scale reconstruction 1pixel-level localization 1swin transformer 1unsupervised anomaly detection 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.CV2026
Semi-Dense Matching Uncertainty Is Not Just Local Confidence
Khoa Hoang, Hoang-Tuan Nguyen, Huong Ninh +2
Reliable semi-dense matching is essential for modern geometric vision systems. Designed under a coarse-to-fine paradigm, it achieves an optimal balance between performance and comp…
cs.CV2026
SwinAD: Multi-stage feature reconstruction for unsupervised industrial anomaly detection
Huong Ninh, Chien Thai, Mai Xuan Trang +3
The paper presents SwinAD, an unsupervised industrial anomaly detection framework that combines a frozen pretrained Swin Transformer V2 encoder with a multi-stage, diversity‑preser…