OmniParser for Pure Vision Based GUI Agent

Yadong Lu1, Jianwei Yang1, Yelong Shen2, Ahmed Awadallah1,
1Microsoft Research, 2Microsoft Gen AI,
Code arXiv Blog post

Abstract

The recent success of large vision language models shows 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 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

Result mobile Result mobile Result mobile

Examples of parsed screenshot image and local semantics by OmniParser. The inputs to OmniParse are user task and UI screenshot, from which it will produce: 1. parsed screenshot image with bounding boxes and numeric IDs overlayed, and 2. local semantics contains both text extracted and icon description.

Curated Dataset for Interactable Region Detection and Icon Functionality Description


We curate a dataset of interactable icon detection dataset, containing 67k unique screenshot images, each labeled with bounding boxes of interactable icons derived from DOM tree. We first took a 100k uniform sample of popular publicly availabe urls on the clueweb dataset, and collect bounding boxes of interactable regions of the webpage from the DOM tree of each urls. We also collected 7k icon-description pairs for finetuning the caption model.
Species Classification results on iWildCam2020-WILDS (OOD) dataset

Examples from the Interactable Region Detection dataset. . TThe bounding boxes are based on the interactable region extracted from the DOM tree of the webpage.

Results


We evaluate our model on SeeClick, Mind2Web, and AITW benchmarks. We show that our model outperforms the GPT-4V baseline on all benchmarks. We also show that our model with screenshot only input outperforms the GPT-4V baselines requiring additional information outside of screenshot.
seeclick mind2web aitw

Plugin-ready for Other Vision Language Models


To further demonstrate OmniParser is a plugin choice for off-the-shelf vision langauge models, we show the performance of OmniParser combined with recently announced vision language models: Phi-3.5-V and Llama-3.2-V. As seen in table, our finetuned interactable region detection (ID) model significantly improves the task performance compared to grounding dino model (w.o. ID) with local semantics across all subcategories for GPT-4V, Phi-3.5-V and Llama-3.2-V. In addition, the local semantics of icon functionality helps significantly with the performance for every vision language model. In the table, LS is short for local semantics of icon functionality, ID is short for the interactable region detection model we finetune. The setting w.o. ID means we replace the ID model with original Grounding DINO model not finetuned on our data, and with local semantics. The setting w.o. ID and w.o LS means we use Grounding DINO model, and further without using the icon description in the text prompt.
seeclick

Demo of Mind2Web Tasks


Citation

@misc{lu2024omniparserpurevisionbased,
                title={OmniParser for Pure Vision Based GUI Agent}, 
                author={Yadong Lu and Jianwei Yang and Yelong Shen and Ahmed Awadallah},
                year={2024},
                eprint={2408.00203},
                archivePrefix={arXiv},
                primaryClass={cs.CV},
                url={https://arxiv.org/abs/2408.00203}, 
          }