SherlockNet

SherlockNet

Using Convolutional Neural Networks to explore and label 400 Years of Book Illustrations by the SherlockNet team, BL Labs competition winners for 2016.

Settings

Now showing 1 - 1 of 1
  • Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Dataset,
    SherlockNet data
    (2017)
    Zhao, Luda
    ;
    Do, Brian
    ;
    Wang, Karen
    Using Convolutional Neural Networks to Explore Over 400 Years of Book Illustrations: Starting from February 2016, as part of the British Library Labs Competition, we embarked on a collaboration with the British Library Labs and the British Museum to tag and caption the entire British Library 1M Collection, a set of 1 million book illustrations scanned from books published between 1500 and 1900. We proposed to use convolutional neural networks (CNNs) to perform this automatic tagging and captioning. In addition, we proposed deeper analysis of temporal trends in these images using the explanatory power provided by neural networks.
      5  5