AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Anomaly Detection of Integrated Circuits Package Substrates Using the Large Vision Model SAIC: Dataset Construction, Methodology, and Application

Ruiyun Yu, Bingyang Guo, Haoyuan Li

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Industrial anomaly detection · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add embedding-based distillation (ablation)SAIC · CPS2D-ADDice / mIoU: improve performance by nearly 15 percentage pointsImproved
add multiple prompts (box and mask) in training (ablation)SAIC · CPS2D-ADDice / mIoUImproved
fine-tune SAM into SAIC with distillationSAM · CPS2D-AD fully supervisedmean Dice: 84.43 -> 88.02Improved
fine-tune SAM into SAIC with distillationSAM · CPS2D-AD fully supervisedmean mIoU: 81.44 -> 87.03Improved
use SAIC instead of ENCNet on other datasetsENCNet · NEU-Seg, Severstal, Magneticmean Dice: 78.29 -> 84.14Improved
add half of the annotated abnormal samples (semi-supervised)Patchcore · CPS2D-ADpixel AUPRO: 0.5300 -> 0.6128Improved
add half of the annotated abnormal samples (semi-supervised)Patchcore · CPS2D-ADimage AP: 0.7781 -> 0.8741Improved
add half of the annotated abnormal samples (semi-supervised)Patchcore · CPS2D-ADpixel AUROC: 0.9051 -> 0.8690Hurt
add 5 shots instead of 1 shotSCCAN · CPS2D-AD few-shotmean mIoU: 36.68 -> 38.52Improved
move from unsupervised to fully supervised use of labelssegmentation models · CPS2D-ADlocalizationImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy