The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/d1f3ca5c-f1e4-48f5-9a04-0564572d2b9c

Wang, Haoran; Khademi, Seyran (2023): Ot & Sien, a dataset to help the development of object detection in children's book illustrations. Version 1. 4TU.ResearchData. dataset. https://doi.org/10.4121/d1f3ca5c-f1e4-48f5-9a04-0564572d2b9c.v1

Dataset

• Artificial Intelligence and Image Processing
• Arts and Leisure
• Cultural Understanding
• Information and Computing Sciences

children books, computer vision, illustrations, imagery content, machine learning, object detection, recognition

CC BY 4.0

RefWorks, BibTeX, Reference Manager, Endnote, DataCite, NLM, DC, CFF

by Haoran Wang , Seyran Khademi

The original dataset is Ot & Sien Dataset (https://lab.kb.nl/dataset/ot-sien-dataset). We corrected mistakes and made it ML-ready.

The purpose of this dataset is to help the development of automatic visual object detection in children's book illustrations. The properties of our dataset are summarized as: 

• The dataset consists of illustrations rather than standard photos. 
• 1452 images with 8241 objects (5.7 per image) are annotated including the category and bounding boxes.
• All images are resized to 416 x 416 with black fitting edges to adapt to the training procedure.
• The dataset follows a natural long-tail property, with some object categories being rare.
• The dataset has imbalanced categories.

• 2023-06-23 first online, published, posted

4TU.ResearchData

*.jpg, *.txt, *.yaml

TU Delft, Faculty of Electrical Engineering, Mathematics & Computer Science (EEMCS)