UICVD: A Computer Vision UI Dataset for Training RPA Agents

Madalina Dicu, Adrian Sterca, Camelia Chira, Radu Orghidan

2024

Abstract

This paper introduces the UICVD Dataset, a novel resource fostering advancements in Robotic Process Automation (RPA) and Computer Vision. The paper focuses on recognizing UI (User Interface) components of a web application which is not as well known as recognizing real objects in images in the field of computer vision. This dataset derives from extensive screen captures within an enterprise application, offering a rare, in-depth look at real-world automation and interface scenarios. For RPA, the UICVD Dataset helps in training the machine model of an RPA agent for recognizing various UI components of the web application which is the target of the automation process. In Computer Vision, it serves as an invaluable tool for identifying and understanding user interface elements, ranging from basic icons to intricate structural details. Designed to support a wide spectrum of research and development initiatives, the UICVD Dataset is positioned as a critical asset for technology advancements in automation and user interface recognition. Its extensive, detailed content and ease of access make it a promising resource for enhancing existing applications and inspiring innovations in RPA and Computer Vision.

Paper Citation

in Harvard Style

Dicu M., Sterca A., Chira C. and Orghidan R. (2024). UICVD: A Computer Vision UI Dataset for Training RPA Agents. In Proceedings of the 19th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE; ISBN 978-989-758-696-5, SciTePress, pages 414-421. DOI: 10.5220/0012632600003687

in Bibtex Style

@conference{enase24,
author={Madalina Dicu and Adrian Sterca and Camelia Chira and Radu Orghidan},
title={UICVD: A Computer Vision UI Dataset for Training RPA Agents},
booktitle={Proceedings of the 19th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE},
year={2024},
pages={414-421},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012632600003687},
isbn={978-989-758-696-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Conference on Evaluation of Novel Approaches to Software Engineering - Volume 1: ENASE
TI - UICVD: A Computer Vision UI Dataset for Training RPA Agents
SN - 978-989-758-696-5
AU - Dicu M.
AU - Sterca A.
AU - Chira C.
AU - Orghidan R.
PY - 2024
SP - 414
EP - 421
DO - 10.5220/0012632600003687
PB - SciTePress