Visual Anomaly (VisA)

Stream data with DDA:

from dagshub.streaming import DagsHubFilesystem

fs = DagsHubFilesystem(".", repo_url="https://dagshub.com/DagsHub-Datasets/visa-dataset")

fs.listdir("s3://amazon-visual-anomaly/VisA_20220922.tar")

Description:

Largest Visual Anomaly detection dataset containing objects from 12 classes in 3 domains across 10,821(9,621 normal and 1,200 anomaly) images. Both image and pixel level annotations are provided.

Contact:

Largest Visual Anomaly detection dataset containing objects from 12 classes in 3 domains across 10,821(9,621 normal and 1,200 anomaly) images. Both image and pixel level annotations are provided.

Update Frequency:

Not updated

Managed By:

https://aws.amazon.com/

Resources:

1. resource:
   • Description: The .tar file for the Visual Anomaly (VisA) dataset
   • ARN: arn:aws:s3:::amazon-visual-anomaly/VisA_20220922.tar
   • Region: us-west-2
   • Type: S3 Bucket
   • Explore: None

Tags:

anomaly detection, classification, segmentation, industrial, fewshot

Publication:

1. publication:
   • Title: SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentation
   • URL: https://arxiv.org/abs/2207.14315
   • AuthorName: Yang Zou, Jongheon Jeong, Latha Pemula, Dongqing Zhang, and Onkar Dabeer