Datasets:

AvoCahDoe
/

invoice-annotated-bbox

license: apache-2.0
language:
  - en
  - de
tags:
  - object-detection
  - document-understanding
  - invoice
  - layout-analysis
  - bounding-box
  - ocr-prep
task_categories:
  - object-detection
size_categories:
  - n<1K
pretty_name: Invoice Layout Annotated BBox Dataset
dataset_info:
  features:
    - name: image
      dtype: image
    - name: image_id
      dtype: int64
    - name: document_id
      dtype: int64
    - name: document_filename
      dtype: string
    - name: page_number
      dtype: int64
    - name: width
      dtype: int64
    - name: height
      dtype: int64
    - name: objects
      struct:
        - name: id
          list: int64
        - name: area
          list: float64
        - name: bbox
          list:
            list: float32
            length: 4
        - name: category
          list:
            class_label:
              names:
                '0': invoice_metadata
                '1': vendor_block
                '2': customer_block
                '3': table_block
                '4': line_item
                '5': summary_block
                '6': payment_block
                '7': Column
  splits:
    - name: train
      num_bytes: 40359783
      num_examples: 76
  download_size: 36093604
  dataset_size: 40359783
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Invoice Layout Annotated BBox Dataset

Manually annotated invoice page images exported from AnnotateEverything, with axis-aligned bounding boxes for 8 document-layout regions. Built for training object detectors (YOLO, DETR, etc.) on invoice macro-structure.

Dataset summary

Property	Value
Pages	76
Documents	1
Source PDF	`train_images.pdf`
Total annotations	771
Avg boxes / page	10.14
Image width range	425 – 2853 px
Image height range	570 – 4096 px
Export date	2026-06-22T19:27:38.375Z
Annotation tool	AnnotateEverything
Project	Invoices (id=2)
On-disk size	~40.3 MB

Classes

ID	Name	Description	Color	Count
0	`invoice_metadata`	Invoice number, date, header metadata	`#f59e0b`	83
1	`vendor_block`	Seller / vendor address block	`#22c55e`	73
2	`customer_block`	Client / customer address block	`#ef4444`	74
3	`table_block`	Line items table region	`#ec4899`	76
4	`line_item`	Individual line item row	`#8b5cf6`	257
5	`summary_block`	Totals / summary section	`#3b82f6`	78
6	`payment_block`	Payment / IBAN block	`#06b6d4`	55
7	`Column`	Table column header row	`#64748b`	75

Directory layout

.
├── README.md                 # This dataset card
├── manifest.json             # Project metadata, label definitions, document index
├── annotations.json          # Flat index of all pages + bboxes
├── documents/
│   └── train_images__doc426/
│       ├── document.json     # Document metadata
│       └── pages/
│           └── page_NNN/
│               ├── image.png
│               └── annotations.json
└── samples/                  # README preview images (JPEG)

Annotation format

Each page entry in annotations.json:

{
  "document_id": 426,
  "document_filename": "train_images.pdf",
  "page_number": 1,
  "relative_image": "documents/train_images__doc426/pages/page_001/image.png",
  "width": 2853,
  "height": 4096,
  "annotations": [
    {
      "layer": "Default",
      "label": "invoice_metadata",
      "bbox": [x_min, y_min, x_max, y_max]
    }
  ]
}

Bounding boxes are absolute pixel coordinates [x_min, y_min, x_max, y_max] in top-left origin.

Sample pages

Page 1 (8 boxes)

Original	Annotated

Page 20 (13 boxes)

Original	Annotated

Page 39 (10 boxes)

Original	Annotated

Page 58 (4 boxes)

Original	Annotated

Page 76 (8 boxes)

Original	Annotated

Page 19 (14 boxes)

Original	Annotated

Label distribution

Label	Count
`line_item`	257
`invoice_metadata`	83
`summary_block`	78
`table_block`	76
`Column`	75
`customer_block (Receiver)`	74
`vendor_block (Sender)`	73
`payment_block`	55

Usage

Load annotations from the Hub

import json
from huggingface_hub import hf_hub_download

ann_path = hf_hub_download("AvoCahDoe/invoice-annotated-bbox", "annotations.json", repo_type="dataset")
with open(ann_path) as f:
    pages = json.load(f)

Convert to YOLO (invoice-extractor)

This dataset is the raw export consumed by prepare_annotated_dataset.py in the invoice-extractor training pipeline.

python scripts/prepare_annotated_dataset.py

Citation

If you use this dataset, please cite the upstream annotation project and note the export format (AnnotateEverything bbox export v1).

License

Apache 2.0