UIPro-AndroidControl-Data-v1

Part of the UIPro GUI-agent training suite (ICCV 2025). This repository packages the AndroidControl source into the unified UIPro instruction-tuning format, with coordinates normalized to a [0, 1000] grid.

Android agent CFT data: intent grounding, text localization, OCR and widget listing from AndroidControl episodes.

Dataset at a glance

Total samples	4,014,396
Valid images	13,603
Avg. samples / image	295.11
Coordinate scale	0–1000
Source dataset	AndroidControl

Samples by task

Task	Count
OCR	2,101,573
TextLoc	1,804,335
WidgetList	80,140
IntentGnd	28,348

Repository file structure

File	Description
AndroidControl_s1000_4014396.json	The dataset: a JSON list of 4,014,396 sample objects (schema below).
AndroidControl_s1000_4014396_sample.json	A small preview slice of the same schema, for quick inspection without downloading everything.
AndroidControl_s1000_4014396_images.zip	All screenshots referenced by the image field, preserving the relative paths stored there.
AndroidControl_s1000_4014396_info.json	Full generation report — per-task counts, image statistics, invalid-element breakdown, and the exact processing config.

Unzip AndroidControl_s1000_4014396_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"
ep_id	AndroidControl episode id the step belongs to.
id	Unique sample identifier, formatted autogui_<dataset>_<task>_<n>. The <task> 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.
step_id	Index of this step within its episode.
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 "<width>x<height>" 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 <image> 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

{
  "id": "autogui_AndroidControl_textloc_15202-3-99",
  "conversations": [
    {
      "from": "human",
      "value": "<image>\nLocate the text \"Finish Type\"  (with bbox)"
    },
    {
      "from": "gpt",
      "value": "(0,398,205,431)"
    }
  ],
  "task_attr": "Finish Type",
  "unnormalized_box": [
    0,
    955,
    221,
    1035
  ],
  "image": "AndroidControl/images/com.android.systemui/15202/3.png",
  "wxh": "1080x2400",
  "ep_id": 15202,
  "step_id": 3
}

Usage

import json, os, zipfile
from PIL import Image
from huggingface_hub import hf_hub_download

repo = "HongxinLi/UIPro-AndroidControl-Data-v1"
samples = json.load(open(hf_hub_download(repo, "AndroidControl_s1000_4014396.json", repo_type="dataset")))

images_zip = hf_hub_download(repo, "AndroidControl_s1000_4014396_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, AndroidControl.

Citation

@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}
}