ShowUI-web

[Github](https://github.com/showlab/ShowUI/tree/main) | [arXiv](https://arxiv.org/abs/2411.17465) | [HF Paper](https://huggingface.co/papers/2411.17465) | [Spaces](https://huggingface.co/spaces/showlab/ShowUI) | [Datasets](https://huggingface.co/datasets/showlab/ShowUI-desktop-8K) | [Quick Start](https://huggingface.co/showlab/ShowUI-2B) **ShowUI-web** is a UI-grounding dataset focused on Web visual element grounding. We developed a parser and collected 22K screenshots, retaining only visual-related elements such as those tagged with ‘Button’ or ‘Checkbox’ by removing static text. After download, please unzip the `image.tar.gz` by `tar -xzf image.tar.gz` to get the image data. If you find our work helpful, please consider citing our paper. ``` @misc{lin2024showui, title={ShowUI: One Vision-Language-Action Model for GUI Visual Agent}, author={Kevin Qinghong Lin and Linjie Li and Difei Gao and Zhengyuan Yang and Shiwei Wu and Zechen Bai and Weixian Lei and Lijuan Wang and Mike Zheng Shou}, year={2024}, eprint={2411.17465}, archivePrefix={arXiv}, primaryClass={cs.CV}, url={https://arxiv.org/abs/2411.17465}, } ```[Github](https://github.com/showlab/ShowUI/tree/main) | [arXiv](https://arxiv.org/abs/2411.17465) | [Hugging Face（HF）论文页](https://huggingface.co/papers/2411.17465) | [Spaces](https://huggingface.co/spaces/showlab/ShowUI) | [数据集](https://huggingface.co/datasets/showlab/ShowUI-desktop-8K) | [快速入门](https://huggingface.co/showlab/ShowUI-2B) **ShowUI-web** 是一款面向网页视觉元素锚定任务的UI锚定（UI-grounding）数据集。 我们研发了专用解析工具，采集了22000张网页截图；通过剔除静态文本，仅保留被标注为「按钮（Button）」与「复选框（Checkbox）」等视觉相关的界面元素。 下载完成后，请执行`tar -xzf image.tar.gz`命令解压该压缩包，以获取图像数据。 若您的研究工作得益于本项目，请考虑引用我们的学术论文。 @misc{lin2024showui, title={ShowUI: One Vision-Language-Action Model for GUI Visual Agent}, author={Kevin Qinghong Lin and Linjie Li and Difei Gao and Zhengyuan Yang and Shiwei Wu and Zechen Bai and Weixian Lei and Lijuan Wang and Mike Zheng Shou}, year={2024}, eprint={2411.17465}, archivePrefix={arXiv}, primaryClass={cs.CV}, url={https://arxiv.org/abs/2411.17465}, }

搜集汇总

背景与挑战

背景概述

ShowUI-web是一个专注于Web界面视觉元素定位的数据集，包含22,000张经过筛选的网页截图，仅保留按钮、复选框等视觉相关元素，剔除了静态文本。数据集大小为43.79GB，采用Apache 2.0许可证，适用于视觉-语言-动作模型的研究。