ArtiFact: Museum Artwork Records with Images and Metadata

ArtiFact is a large-scale multimodal benchmark combining artwork records from the Rijksmuseum, the Metropolitan Museum of Art, and the Art Institute of Chicago. The dataset contains aligned images and structured metadata, normalized across fields for artists, dates, materials, and techniques. It was created by deem-data and last updated on June 8, 2026.

Use Cases

• Evaluating metadata extraction models based on structured fields like artist, date, and materials.
• Training error detection systems on normalized museum catalog records.
• Benchmarking semantic querying capabilities over multimodal cultural-heritage collections.
• Developing multimodal reasoning models using aligned images and textual metadata.

Strengths

• Combines records from three major art institutions: the Rijksmuseum, the Metropolitan Museum of Art, and the Art Institute of Chicago.
• Includes aligned images and structured metadata, enabling multimodal tasks.
• Features normalized fields for artists, dates, materials, and techniques.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is from the future (2026).

Provenance

Source

Rijksmuseum, Metropolitan Museum of Art (Met), Art Institute of Chicago (AIC)

Collection Method

Combined and normalized records from the three museum collections.

Freshness

Last updated 2026-06-08 13:32:22