Skip to content

Dataset · gigant

Scans of vintage book illustrations with descriptions

gigant/oldbookillustrations

A collection of 4,172 scanned engravings and drawings from old books with text captions for image generation and analysis.

What's inside

The dataset contains 4,172 scanned illustrations from antique books along with captions and descriptions. Descriptions are primarily in English with occasional fragments in French and German. The materials were gathered from the Old Book Illustrations website with the owner's permission as part of the BigLAM initiative.

What you can do with it

  • Training LoRA or stylized text-to-image models in the aesthetic of book engravings
  • Training image-to-text models to describe vintage illustrations
Tasks
Text → image, Image → text, Image → image
Languages
English, French, German
Size
1K–10K rows
Downloads this month
1.6K
Likes
41

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("gigant/oldbookillustrations")