Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs

TL;DR

Ferret-UI is a specialized multimodal large language model designed for detailed understanding and interaction with mobile UI screens, outperforming existing models and GPT-4V on various UI tasks.

Contribution

The paper introduces Ferret-UI, a tailored MLLM for mobile UI understanding with novel multi-resolution encoding and extensive UI-specific training datasets.

Findings

Ferret-UI surpasses most open-source UI MLLMs in comprehension.

Ferret-UI outperforms GPT-4V on elementary UI tasks.

The model demonstrates strong reasoning and interaction capabilities.

Abstract

Recent advancements in multimodal large language models (MLLMs) have been noteworthy, yet, these general-domain MLLMs often fall short in their ability to comprehend and interact effectively with user interface (UI) screens. In this paper, we present Ferret-UI, a new MLLM tailored for enhanced understanding of mobile UI screens, equipped with referring, grounding, and reasoning capabilities. Given that UI screens typically exhibit a more elongated aspect ratio and contain smaller objects of interest (e.g., icons, texts) than natural images, we incorporate "any resolution" on top of Ferret to magnify details and leverage enhanced visual features. Specifically, each screen is divided into 2 sub-images based on the original aspect ratio (i.e., horizontal division for portrait screens and vertical division for landscape screens). Both sub-images are encoded separately before being sent to LLMs. We meticulously gather training samples from an extensive range of elementary UI tasks, such as icon recognition, find text, and widget listing. These samples are formatted for instruction-following with region annotations to facilitate precise referring and grounding. To augment the model's reasoning ability, we further compile a dataset for advanced tasks, including detailed description, perception/interaction conversations, and function inference. After training on the curated datasets, Ferret-UI exhibits outstanding comprehension of UI screens and the capability to execute open-ended instructions. For model evaluation, we establish a comprehensive benchmark encompassing all the aforementioned tasks. Ferret-UI excels not only beyond most open-source UI MLLMs, but also surpasses GPT-4V on all the elementary UI tasks.