Skip to content

tzjtatata/SCAN

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 

Repository files navigation

Triad: Empowering LMM-based Anomaly Detection with Vision Expert-guided Visual Tokenizer and Manufacturing Process.

Beyond Common Sense: Grounding Logical Anomaly Detection in Inspection Criteria
Yuanze Li*, Shihao Yuan, Zimeng Zhu, Ming Liu (csmliu@outlook.com), Wangmeng Zuo, Guangming Shi

TODO

  • update Triad model pretrained weights.
  • upload SCAN Training and Evaluation codes.
  • update evaluation benchmark.
  • update demo.

Contents

Install

Coming soon.

Triad Weights

Coming soon.

Demo

Coming soon.

Train

Coming soon.

Evaluation

Coming soon.

Acknowledgement

Thanks to Qwen Team for their great base model Qwen2.5-VL and Qwen3-VL series.

About

The official implementation of the paper "Beyond Common Sense: Grounding Logical Anomaly Detection in Inspection Criteria" which is accepted by ECCV 26. This work introduces criteria for logical anomaly detection and provide a new large-scale training set for LAD, FLAW.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors