Style-Aware Sketch-to-Code Conversion for the Web
Abstract
While sketching a graphical user interface (GUI) is a necessary step towards the creation of a Web application, its transformation into a coded GUI, with the proper styles, is still a tedious and time-consuming task that a designer should perform. Recently, a set of Machine Learning techniques has been applied to automatically generate code from sketches to ease this part of the design process. These techniques effectively convert the sketches into a skeleton structure of the GUI but are not designed to consider the styles to be applied to the generated HTML page. Moreover, having the possibility to explore different styles, starting from a sketch, might further support the designer in their work. In this paper, we move the first steps to enable this opportunity by proposing a method that allows the designer to input the sketch of the Web-based GUI and select a reference style to be applied. Our method automatically injects the reference styles into the sketch components and then uses an automatic code generation technique to obtain the final code. Preliminary experiments carried out with the navigation bar component show the effectiveness of the proposed method.
BibTeX
@inproceedings{10.1145/3531706.3536462,
author = {Cal\`{o}, Tommaso and De Russis, Luigi},
title = {Style-Aware Sketch-to-Code Conversion for the Web},
year = {2022},
isbn = {9781450390316},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3531706.3536462},
doi = {10.1145/3531706.3536462},
pages = {44–47},
numpages = {4},
keywords = {convolutional neural network, machine learning, user interface, web elements},
location = {Sophia Antipolis, France},
series = {EICS '22 Companion}
}