OmniParser for Pure Vision Based GUI Agent
Lists
Only you see this
Abstract
The recent advancements of large vision language models shows their great potential in driving the agent system operating on user interfaces. However, we argue that the power multimodal models like GPT-4V as a general agent on multiple operating systems across different applications is largely underestimated due to the lack of a robust screen parsing technique capable of: 1) reliably identifying interactable icons within the user interface, and 2) understanding the semantics of various elements in a screenshot and accurately associate the intended action with the corresponding region on the screen. To fill these gaps, we introduce OmniParser, a comprehensive method for parsing general user interface screenshots into structured elements, which significantly enhances the ability of GPT-4V to generate actions that can be accurately grounded in the corresponding regions of the interface. We first curated an interactable icon detection dataset using popular webpages and an icon description dataset. These datasets were utilized to fine-tune specialized models: a detection model to parse interactable regions on the screen and a caption model to extract the functional semantics of the detected elements. OmniParser significantly improves GPT-4V's performance on ScreenSpot benchmark. And on Mind2Web and AITW benchmark, OmniParser with screenshot only input outperforms the GPT-4V baselines requiring additional information outside of screenshot. We further demonstrate that OmniParser can seamlessly integrate with other vision language models, significantly enhancing their agentic capabilities.
BibTeX
@misc{lu2024omniparser,
author = {Yadong Lu and Jianwei Yang and yelong shen and Ahmed Hassan Awadallah},
title = {{OmniParser for Pure Vision Based GUI Agent}},
year = {2024},
eprint = {2408.00203},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2408.00203},
}