This repository is for our AAAI'23 paper:
Layout Representation Learning with Spatial and Structural Hierarchies PDF
Yue Bai, Dipu Manandhar, Zhaowen Wang, John Collomosse, and Yun Fu
This paper proposes a spatial-structural auto-encoder for layout representation learning. The layout element patterns are considered from both spatial and structural perspectives in a hierarchical modeling fashion. Correspondingly, we propose to use a new tree-edit distance (TED) based metric to evaluate the layout representation quality from structural aspect, which further provides a comprehensive evaluation protocol. In addition, we newly collect a POSTER dataset containing diverse design template for future research work of layout related topics.
We refer to the following repository for our implementations: Learning Structural Similarity of User Interface Layouts using Graph Networks. We appreciate their great works!
Please cite this in your publication if our work helps your research. Should you have any questions, welcome to reach out to Yue Bai (bai.yue@northeastern.edu).
@article{bai2023layout,
title={Layout Representation Learning with Spatial and Structural Hierarchies},
author={Bai, Yue and Manandhar, Dipu and Wang, Zhaowen and Collomosse, John and Fu, Yun},
year={2023}
}