Form Understanding Dataset with Region Annotations

Serving as derived from the FUNSD benchmark for form understanding. It contains annotated form images with labeled regions. The dataset was uploaded by user sijpapi and last updated in September 2021.

Use Cases

• Train a layout analysis model to detect and classify Regionus annotations on form images.
• Fine-tune a document understanding model using the region-level annotations for information extraction.
• Benchmark form parsing algorithms against the FUNSD benchmark's region annotation schema.

Strengths

• Derived from the established FUNSD benchmark for form understanding.
• Includes region-level annotations (Regionus tag) for structured document analysis.
• Last updated in September 2021, providing a historical snapshot.

Limitations

• The specific number of rows, columns, and sample data is unavailable, limiting assessment of scale and structure.
• The dataset's connection to the original FUNSD benchmark is not detailed, potentially obscuring its exact composition.
• No license information is provided, which restricts clarity on permissible use.

Provenance

Source

Derived from the FUNSD (Form Understanding in Noisy Scanned Documents) dataset.

Collection Method

Uploaded by user sijpapi; specific collection or processing method is unknown.

Freshness

Last updated 2021-09-07.