A prototype-aligned multi-scale attention network for few-shot industrial surface defect detection

Rubing Huang, Lianyi Zhao · 2026

Read A prototype-aligned multi-scale attention network for few-shot industrial surface defect detection by Rubing Huang, Lianyi Zhao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0359400.txt.

Description

Huang and one co-author published a 2026 PLOS ONE paper on finding flaws on factory surfaces with very few examples. Their network learns from a handful of defect photos by matching new images against stored prototypes. Attention at several scales helps it catch both tiny and large flaws. Inspectors often have few samples of rare defects, so methods that learn from little are valuable.

Catalog record

CREATORRubing Huang, Lianyi Zhao
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE76188
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0359400

Browse the full library · About the Tower of Babel