Datasets:

movensys
/

cube-detection-monoply-background-obb

license: mit
task_categories:
  - object-detection
tags:
  - yolo
  - obb
  - oriented-bounding-box
  - cubes
  - robotics
  - pick-and-place
pretty_name: Cube Detection on Monopoly Board Background (OBB)
size_categories:
  - n<1K

Cube Detection on Monopoly Board Background (OBB)

A small oriented-bounding-box (OBB) detection dataset of colored cubes placed on a Movensys "Monopoly" board background. Intended for fine-tuning YOLO-style OBB detectors used in pick-and-place / robotic manipulation pipelines.

Classes

ID	Name
0	green_cube
1	yellow_cube
2	blue_cube
3	red_cube

Splits

Split	Images	Labels
train	104	104
val	29	29
test	16	16
total	149	149

Image format

Resolution: 1280 × 720, RGB JPEG
Captured from a top-down camera over a printed Movensys Monopoly board, with colored cubes placed at varying positions and orientations

Label format

YOLO OBB — one row per object, 9 values:

class_id x1 y1 x2 y2 x3 y3 x4 y4

All polygon coordinates are normalized to [0, 1] relative to image width/height. Vertices are given in order around the box.

Example (train/labels/00001.txt):

0 0.0522 0.1119 0.1013 0.0214 0.1529 0.1101 0.1038 0.2005
3 0.2423 0.0615 0.3122 0.0615 0.3122 0.1869 0.2423 0.1869

Directory layout

.
├── dataset.yaml
├── train/
│   ├── images/   # *.jpg
│   └── labels/   # *.txt
├── val/
│   ├── images/
│   └── labels/
└── test/
    ├── images/
    └── labels/

Usage

Download

hf download movensys/cube-detection-monoply-background-obb \
    --repo-type dataset \
    --local-dir ./cube-detection-monoply-background-obb

Train with Ultralytics YOLO (OBB)

After download, update the path: field in dataset.yaml to point at the local copy:

path: /absolute/path/to/cube-detection-monoply-background-obb
train: train/images
val: val/images
test: test/images
names:
  0: green_cube
  1: yellow_cube
  2: blue_cube
  3: red_cube

Then:

from ultralytics import YOLO

model = YOLO("yolo11n-obb.pt")
model.train(data="dataset.yaml", epochs=100, imgsz=1280)

License

Released under the MIT License.