MVTec Anomaly Detection Industrial Images

5,354 industrial product images for anomaly detection, originally from MVTec and provided by Voxel51. The dataset is formatted for loading with the FiftyOne library, facilitating computer vision tasks in manufacturing inspection.

Use Cases

• Train anomaly detection models on industrial images to identify defects in manufacturing products.
• Benchmark computer vision algorithms for industrial inspection using the standardized MVTec AD format.
• Load and visualize image data with the FiftyOne library for rapid prototyping and analysis.

Strengths

• Contains 5,354 total samples, matching the scale of the original MVTec AD benchmark.
• Sourced from the established MVTec AD dataset, a standard for industrial anomaly detection research.
• Formatted for easy integration with the FiftyOne computer vision toolkit.

Limitations

• Specific column definitions, file formats, and image resolutions are not detailed in the provided input.
• The dataset's composition (e.g., number of object categories, train/test split) is not specified here.
• The last known update was in January 2025, but the original data collection timeframe is unknown.

Provenance

Source

MVTec, provided by Voxel51.

Collection Method

Originally created by MVTec for anomaly detection research, reformatted for FiftyOne.

Freshness

Last updated on 2025-01-30.