README.md

metadata

task_categories:
  - object-detection
license: cc-by-nc-4.0
tags:
  - Materials Informatics
  - AI for Science
dataset_info:
  features:
    - name: image
      dtype: image
    - name: image_id
      dtype: int64
    - name: width
      dtype: int64
    - name: height
      dtype: int64
    - name: bbox
      list:
        list: float64
    - name: category
      list: string
    - name: category_id
      list: int64
    - name: area
      list: float64
    - name: iscrowd
      list: int64
  splits:
    - name: train
      num_bytes: 2518645352
      num_examples: 2855
  download_size: 2321915601
  dataset_size: 2518645352
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
pretty_name: MaterialScope
arxiv: 2606.29667

MaterialScope

A manually annotated object detection dataset for compound material image to sub-image region detection, in COCO format.
All 2,855 images are fully annotated — no unannotated images in the split.

---

Dataset Summary

Metric	Value
Total images	2,855
Total annotations	13,008
Categories	23
Avg. annotations per image	4
Max annotations per image	21
Annotation format	COCO JSON
Image format	JPG

---

Category Distribution

Category	Annotation Count
A	2,552
B	2,539
C	2,192
D	1,979
E	1,145
F	867
G	436
H	346
I	219
J	109
K	85
L	64
M	33
N	26
O	19
P	8
Q	5
R	4
S	3
T	3
single	272
common	41
unlabel	61
Total	13,008

---

File Structure

merged_coco.json   ← COCO annotation file (images + annotations + categories)
images/            ← 2,855 JPG images

---

Annotation Format

Annotations follow the standard COCO format:

{
  "images": [
    { "id": 1, "file_name": "img1.jpg", "width": 1721, "height": 781 }
  ],
  "annotations": [
    {
      "id": 1,
      "image_id": 1,
      "category_id": 21,
      "bbox": [x, y, width, height],
      "area": 1318264.48,
      "iscrowd": 0
    }
  ],
  "categories": [
    { "id": 1, "name": "A" },
    { "id": 2, "name": "B" },
    ...
  ]
}

Bounding boxes are in [x_min, y_min, width, height] format (COCO standard).

---

Loading the Dataset

With Python (raw COCO)

import json
from PIL import Image

with open("merged_coco.json") as f:
    coco = json.load(f)

# Build a lookup: image_id → annotations
from collections import defaultdict
ann_by_image = defaultdict(list)
for ann in coco["annotations"]:
    ann_by_image[ann["image_id"]].append(ann)

# Load an image and its annotations
img_info = coco["images"][0]
image    = Image.open(f"images/{img_info['file_name']}")
anns     = ann_by_image[img_info["id"]]
print(f"{img_info['file_name']}: {len(anns)} annotations")

With pycocotools

from pycocotools.coco import COCO

coco = COCO("merged_coco.json")
img_ids = coco.getImgIds()
ann_ids = coco.getAnnIds(imgIds=img_ids[0])
anns    = coco.loadAnns(ann_ids)

With Ultralytics (YOLO training)

pip install ultralytics

from ultralytics.data.converter import convert_coco

convert_coco(
    labels_dir=".",
    save_dir="yolo_dataset",
    use_segments=False,
)

---

License

This dataset is licensed under Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0).

• Free to use for research and non-commercial purposes
• You must give appropriate credit when using or sharing this dataset
• Derivatives must be shared under the same license
• Commercial use is not permitted

Full license text: https://creativecommons.org/licenses/by-nc/4.0/

---

Citation

If you use this dataset in your research, please cite it as:

@article{ghosh2026unlocking,
  title={Unlocking the Visual Record of Materials Science: A Large-Scale Multimodal Dataset from Scientific Literature},
  author={Ghosh, Subham and Tiwari, Shubham and Ibrahim, Mohammad and Tewari, Abhishek},
  journal={arXiv preprint arXiv:2606.29667},
  year={2026},
  doi={https://doi.org/10.48550/arXiv.2606.29667}
}