|
|
--- |
|
|
dataset_info: |
|
|
features: |
|
|
- name: image |
|
|
dtype: image |
|
|
- name: label |
|
|
dtype: string |
|
|
splits: |
|
|
- name: train |
|
|
num_bytes: 65880607.6 |
|
|
num_examples: 16 |
|
|
- name: test |
|
|
num_bytes: 15634112.4 |
|
|
num_examples: 4 |
|
|
download_size: 81521051 |
|
|
dataset_size: 81514720 |
|
|
configs: |
|
|
- config_name: default |
|
|
data_files: |
|
|
- split: train |
|
|
path: data/train-* |
|
|
- split: test |
|
|
path: data/test-* |
|
|
license: apache-2.0 |
|
|
task_categories: |
|
|
- object-detection |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- objectdetection d |
|
|
- detection |
|
|
- syntheticdata |
|
|
- yolov8 |
|
|
- yolo |
|
|
- labels |
|
|
- labeled |
|
|
- label |
|
|
- indoor |
|
|
- cpg |
|
|
- can |
|
|
size_categories: |
|
|
- 1K<n<10K |
|
|
--- |
|
|
Soup Can Object Detection Dataset Sample |
|
|
Duality.ai just released a 1000 image dataset used to train a YOLOv8 model for object detection -- and it's 100% free! |
|
|
Just [create an EDU account here](link). |
|
|
This HuggingFace dataset is a 20 image and label sample, but you can get the rest at no cost by [creating a FalconCloud account](link). Once you verify your email, the link will redirect you to the dataset page. |
|
|
|
|
|
What makes this dataset unique, useful, and capable of bridging the Sim2Real gap? |
|
|
- The digital twins are not generated by AI, but instead crafted by 3D artists to be INDISTINGUISHABLE to the model from the physical-world objects. This allows the training from this data to transfer into real-world applicability |
|
|
- The simulation software, called FalconEditor, can easily create thousands of images with varying lighting, posing, occlusions, backgrounds, camera positions, and more. This enables robust model training. |
|
|
- The labels are created along with the data. This not only saves large amounts of time, but also ensures the labels are incredibly accurate and reliable. |
|
|
|
|
|
|
|
|
Dataset Overview |
|
|
This dataset consists of high-quality images of soup cans captured in various poses and lighting conditions .This dataset is structured to train and test object detection models, specifically YOLO-based and other object detection frameworks. |
|
|
Why Use This Dataset? |
|
|
Single Object Detection: Specifically curated for detecting soup cans, making it ideal for fine-tuning models for retail, inventory management, or robotics applications. |
|
|
|
|
|
Varied Environments: The dataset contains images with different lighting conditions, poses, and occlusions to help solve traditional recall problems in real world object detection. |
|
|
|
|
|
Accurate Annotations: Bounding box annotations are precise and automatically labeled in YOLO format as the data is created. |
|
|
Create your own specialized data! |
|
|
You can create a dataset like this but with your own digital twin! [Create an account and follow this tutorial to learn how](link). |
|
|
Dataset Structure |
|
|
The dataset is organized as follows: |
|
|
Soup-Can-Object-Detection-Dataset/ |
|
|
│-- images/ |
|
|
│ ├── 000000000.png |
|
|
│ ├── 000000001.png |
|
|
│ ├── ... |
|
|
│-- labels/ |
|
|
│ ├── 000000000.txt |
|
|
│ ├── 000000001.txt |
|
|
│ ├── ... |
|
|
|
|
|
Components |
|
|
Images: RGB images of the soup can in .png format. |
|
|
Labels: .txt files containing bounding box annotations in the YOLO format. |
|
|
0 = soup can |
|
|
Example Annotation (YOLO Format): |
|
|
|
|
|
0 0.475 0.554 0.050 0.050 |
|
|
|
|
|
Where: |
|
|
0 represents the object class (soup can). |
|
|
The next four values represent the bounding box coordinates (normalized x_center, y_center, width, height). |
|
|
Usage |
|
|
This dataset is designed to be used with popular deep learning frameworks: |
|
|
|
|
|
from datasets import load_dataset |
|
|
|
|
|
dataset = load_dataset("your-huggingface-username/Soup-Can-Object-Detection") |
|
|
|
|
|
To train a YOLOv8 model, you can use Ultralytics' yolo package: |
|
|
|
|
|
yolo train model=yolov8n.pt data=soup_can.yaml epochs=50 imgsz=640 |
|
|
|
|
|
|
|
|
Licensing |
|
|
License: Apache 2.0 |
|
|
Attribution: If you use this dataset in research or commercial projects, please provide appropriate credit. |
