PRISM: Moire-Contaminated Fabric Images with Clean Targets

A computer vision dataset for image restoration, released by author running-Turtle. It contains pairs of moire-contaminated fabric images and their corresponding clean ground truth images, organized into training, test, and real-world-image splits. The dataset page was last updated on 2026-06-22.

Use Cases

• Train image denoising models based on paired moire-contaminated and clean fabric images.
• Benchmark moire removal algorithms using the provided test split.
• Evaluate model generalization on real-world moire patterns using the 'real-world-image' subset.

Strengths

• Dataset structure is explicitly defined with separate training, test, and real-world-image directories.
• Provides paired data (input and ground truth) which is essential for supervised learning tasks.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

huggingface

Freshness

Last updated 2026-06-22 09:13:32; freshness should be verified.