--- license: cc-by-nc-4.0 task_categories: - image-to-text - visual-question-answering language: - en tags: - gui - gui-agent - ui-understanding - screenshot - visual-grounding pretty_name: UIPro WebUI size_categories: - 100K Web-page screenshots with icon grounding/referring, element classification, intent grounding and widget listing. ## Dataset at a glance | | | | :--- | :--- | | **Total samples** | 488,499 | | **Valid images** | 327,280 | | **Avg. samples / image** | 1.49 | | **Coordinate scale** | 0–1000 | | **Source dataset** | [WebUI](https://huggingface.co/datasets/biglab/webui-all) | ### Samples by task | Task | Count | | :--- | ---: | | WidgetList | 160,266 | | IconRef | 116,403 | | IntentGnd | 110,800 | | IconGnd | 59,561 | | ElemClass | 41,469 | ## Repository file structure | File | Description | | :--- | :--- | | `WebUI_scale1000_488k.json` | The dataset: a JSON **list** of 488,499 sample objects (schema below). | | `WebUI_scale1000_488k_sample.json` | A small preview slice of the same schema, for quick inspection without downloading everything. | | `WebUI_scale1000_488k_images.zip` | All screenshots referenced by the `image` field, preserving the relative paths stored there. | | `WebUI_scale1000_488k_info.json` | Full generation report — per-task counts, image statistics, invalid-element breakdown, and the exact processing config. | Unzip `WebUI_scale1000_488k_images.zip` and each sample's `image` path resolves relative to the extraction root. ## Sample schema — every field explained Each element of the main JSON list is one training sample. This dataset's samples use the following fields: | Field | Meaning | | :--- | :--- | | `conversations` | The vision-language dialogue: a list of turns, each `{"from": "human"|"gpt", "value": ...}`. The **human** turn holds the instruction/question and contains the `` placeholder marking where the screenshot is inserted; the **gpt** turn is the ground-truth answer. | | `id` | Unique sample identifier, formatted `autogui___`. The `` segment (e.g. `intentgnd`, `textloc`, `ocr`, `elemgnd`, `elemref`) tells you which task the sample belongs to. | | `image` | Path to the screenshot **inside `_images.zip`**, relative to the archive root. Load the image by joining this path with your extraction directory. | | `task_attr` | The task's target attribute in plain form — for grounding tasks the referred element's text/instruction; for OCR/referring tasks the queried coordinate string. Useful for filtering or building custom prompts without parsing the conversation. | | `unnormalized_box` | Ground-truth bounding box in **original image pixels**, as `[x1, y1, x2, y2]` (top-left, bottom-right). Present when a box is available. Note: the answer in the `gpt` turn is **normalized to 0–1000**, while this field is the raw-pixel box — divide by width/height and multiply by 1000 to reconcile them. | | `wxh` | Original screenshot size as the string `"x"` in pixels — use it to convert between the normalized 0–1000 coordinates and raw pixels. | > **Note:** Fields tied to a bounding box (e.g. `unnormalized_box`) are only present on samples > that have a box; point-only answers (e.g. some intent-grounding samples) may omit them. ### The `conversations` field in detail `conversations` is a list of turns that a vision-language model consumes directly: - Each turn is `{"from": "...", "value": "..."}`. - `from` is either **`human`** (the prompt) or **`gpt`** (the ground-truth response). - The token **``** inside a human turn marks where the screenshot is spliced into the prompt — replace it with the actual image when tokenizing. ### Coordinate system - Answers are **normalized to the `0–1000` range** relative to image width/height. - A **point** answer is formatted `(x,y)`; a **bounding box** answer is `(x1,y1,x2,y2)`. - Prompts ending in `(with point)` expect a point; `(with bbox)` expect a box. - `unnormalized_box`, when present, is the same box in **raw pixels** — combine it with `wxh` (or the image's true size) to convert between pixels and the normalized grid. ## Example ```json { "id": "autogui_webui_widgetlist_107976", "conversations": [ { "from": "human", "value": "\nPlease list all the elements on the UI screen." }, { "from": "gpt", "value": "0 StaticText '3个月考上北大法硕，她说普通人成功很简单' (26,14,708,64)\n1 link '' (731,14,972,90)\n2 StaticText '时间：2022-07-20 18:11' (26,71,405,88)\n3 StaticText '老鼠吓退大猫抢吃猫粮 猫连连后退拱手相让' (26,121,692,171)\n4 link '' (731,121,972,197)\n5 StaticText '时间：2022-07-20 17:32' (26,178,405,194)\n6 StaticText '普通人镜头下的真生活' (26,227,462,251)\n7 link '' (731,227,972,303)\n8 StaticText '时间：2022-07-20 16:15' (26,284,405,301)\n9 StaticText '加载更多' (413,333,587,355)\n10 emphasis '' (26,416,33,440)\n11 StaticText '更多' (887,418,959,437)\n12 StaticText '吴亦凡打卡的面馆被整改' (26,461,505,485)\n13 link '' (731,461,972,537)\n14 StaticText '时间：2022-07-20 18:34' (26,518,405,534)\n15 StaticText '广州人才入户年龄放宽5年 本科40岁硕士45岁博士50岁' (26,568,690,617)\n16 link '' (731,568,972,643)\n17 StaticText '时间：2022-07-20 17:58' (26,624,405,641)\n18 StaticText '兵马俑手指在美被折断盗走' (26,674,549,698)\n19 link '' (731,674,972,750)\n20 StaticText '时间：2022-07-20 17:03' (26,731,405,748)\n21 StaticText '凤凰周刊：如果不靠高考，你该如何体面地出人头..' (26,781,679,831)\n22 link '' (731,781,972,857)\n23 StaticText '时间：2022-07-20 17:03' (26,838,405,854)\n24 StaticText '王中磊夫妇为儿子庆生 威廉弟弟长高了许多' (26,887,692,937)\n25 link '' (731,887,972,963)\n26 StaticText '时间：2022-07-20 16:16' (26,944,405,961)" } ], "unnormalized_box": [ [ 30, 36, 828, 162 ], [ 855, 36, 1137, 228 ], [ 30, 180, 474, 22 ... (truncated) ``` ## Usage ```python import json, os, zipfile from PIL import Image from huggingface_hub import hf_hub_download repo = "HongxinLi/UIPro-WebUI-Data-v1" samples = json.load(open(hf_hub_download(repo, "WebUI_scale1000_488k.json", repo_type="dataset"))) images_zip = hf_hub_download(repo, "WebUI_scale1000_488k_images.zip", repo_type="dataset") with zipfile.ZipFile(images_zip) as zf: zf.extractall("images/") s = samples[0] print(s["conversations"]) img = Image.open(os.path.join("images", s["image"])) # screenshot for this sample print("image size:", img.size) ``` ## About UIPro UIPro is a generalist GUI agent trained on 20.6M understanding tasks across 13 task types, followed by agent continued fine-tuning. See the project repository for the full data pipeline, training recipes and evaluation scripts: **https://github.com/ZJULiHongxin/UIPro** ## License Released under **CC BY-NC 4.0** (non-commercial research use). The underlying screenshots and annotations remain subject to the terms of their original source, [WebUI](https://huggingface.co/datasets/biglab/webui-all). ## Citation ```bibtex @inproceedings{uipro2025, title = {UIPro: A Generalist GUI Agent}, author = {Li, Hongxin and others}, booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}, year = {2025} } ```