PubLayNet: Document Layout Detection Data in YOLO and COCO Formats

A dataset for document layout analysis, likely containing annotated images of academic papers. The data is formatted for object detection models, specifically in YOLO and COCO formats, and is split into training, validation, and test sets. It is hosted on Kaggle, but the original author, organization, and specific scale are unknown.

Use Cases

• Train an object detection model to identify document regions like text, titles, and figures (inferred from domain, verify after download)
• Benchmark layout detection algorithms using standardized COCO metrics (inferred from domain, verify after download)
• Fine-tune a YOLO model for a custom document processing pipeline (inferred from domain, verify after download)

Strengths

• Published on Kaggle, a major platform for data science resources.
• Data is pre-split into training, validation, and test partitions for machine learning.

Limitations

• Metadata is minimal; actual content, scale, and annotation quality require verification after download.
• Column-level documentation and sample data are unavailable, making field semantics unclear.
• The license, last update date, and original author are unknown, affecting reproducibility and trust.

Provenance

Source

Kaggle

Collection Method

Likely derived from the original PubLayNet dataset for document layout detection.

Time Range

null

Freshness

Last updated date is unknown; freshness unverified.

Geography

null