mala-lab/deviation-network-image

Official PyTorch implementation of the paper “Explainable Deep Few-shot Anomaly Detection with Deviation Networks”, weakly/partially supervised anomaly detection, few-shot anomaly detection, image defect detection.

34
/ 100
Emerging

This tool helps quality control inspectors and manufacturing engineers automatically identify defects in products using images. You provide a small set of images of good products and a few examples of known defects, and it learns to spot anomalies in new images. The output highlights where the defects are located, making it easier to inspect goods on an assembly line or in a quality assurance process.

No commits in the last 6 months.

Use this if you need to detect unusual patterns or defects in image data, especially when you have very few examples of what constitutes a 'defect'.

Not ideal if your data is not image-based, such as numerical tables or text, or if you have a very large, well-labeled dataset of both normal and anomalous items.

quality-control manufacturing-inspection defect-detection visual-inspection anomaly-detection
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 17 / 25

How are scores calculated?

Stars

88

Forks

15

Language

Python

License

Last pushed

Oct 29, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/mala-lab/deviation-network-image"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.