Automated Cross-Platform GUI Code Generation for Mobile Apps
Authors:
- 1 February 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 13-16
Abstract
Android and iOS are the two dominant platforms for building mobile apps. To provide uniform and smooth user experience, app companies typically employ two teams of programmers to develop UIs (and underlying functionalities) for these two platforms, respectively. However, this development practice is costly for both development and maintenance. To reduce the cost, we take the first step in this direction by proposing an automated cross-platform GUI code generation framework. It can transfer the GUI code implementation between the two mobile platforms. Specifically, our framework takes as input the UI pages and outputs the GUI code for the target platform (e.g., Android or iOS). It contains three phases, i.e., component identification, component type mapping, and GUI code generation. It leverages image processing and deep learning classification techniques. Apart from the UI pages of an app, this framework does not require any other inputs, which makes it possible for large-scale, platform -independent code generation.Keywords
This publication has 9 references indexed in Scilit:
- Efficiently manifesting asynchronous programming errors in Android apps2018
- pix2code2018
- Large-scale analysis of framework-specific exceptions in Android apps2018
- From UI design image to GUI skeleton2018
- Guided, stochastic model-based GUI testing of Android apps2017
- FSMdroid2016
- Reverse Engineering Mobile Application User Interfaces with REMAUI (T)2015
- An Exploratory Study on the Relation between User Interface Complexity and the Perceived Quality2014
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998