Scene text editing (STE) is a challenging task due to a complex intervention between text and style . We propose a novel representational learning-based STE model that employs textual information as well as visual information . Our experiments demonstrate that RewriteNet achieves betterquantitative and qualitative performance than other comparisons . Wevalidate that the use of text information and the self-supervised trainingscheme improves text switching performance . The implementation and dataset will be publicly available. The implementation of the STE model and dataset is publicly available to the public, with the details of the implementation and data will be available . Back to the page you came from, contact us at http://www.mailonline.co.uk/newslink/researches/recode-networks/recovery/recovering-recoquetry-training-network-recoverable-model-back-to-the-text-changing-researsing-search-reusing-thetext-change-change

Author(s) : Junyeop Lee, Yoonsik Kim, Seonghyeon Kim, Moonbin Yim, Seung Shin, Gayoung Lee, Sungrae Park

Links : PDF - Abstract

Code :
Coursera

Keywords : text - implementation - model - information - ste -

Leave a Reply

Your email address will not be published. Required fields are marked *