Hex: Lego Computer Vision Project
Updated 5 months ago
416
views35
downloadsMetrics
Overview The Hex: Lego Object Detection Model utilizes the YOLOv7 algorithm to accurately identify and classify various sizes and colors of LEGO bricks. This model is designed to perform robustly under diverse lighting conditions, suitable for applications in automated sorting, inventory management, and educational tools.
Dataset This dataset, curated with Roboflow, includes 8,320 images and over 15,000 annotations, capturing LEGO bricks in different configurations and lighting environments. Data augmentation techniques were applied to enhance robustness, covering 28 distinct LEGO brick classes.
Model Using the YOLOv7 architecture, the model achieves high precision with a mean Average Precision (mAP) of 88.3% and a precision rate of 95.6%. It supports real-time detection, making it ideal for dynamic applications.
Training and Validation The model was first trained using gpu and then uploaded to roboflow to refine, with extensive hyperparameter tuning. Validation tests in both controlled and dynamic environments ensured comprehensive performance assessment.
Applications The model is suitable for:
- Automated Sorting: Enhancing robotic systems for LEGO sorting.
- Inventory Management: Assisting in tracking LEGO inventories.
- Educational Tools: Providing interactive learning through real-time LEGO detection.
Use This Trained Model
Try it in your browser, or deploy via our Hosted Inference API and other deployment methods.
Build Computer Vision Applications Faster with Supervision
Visualize and process your model results with our reusable computer vision tools.
Cite This Project
If you use this dataset in a research paper, please cite it using the following BibTeX:
@misc{
hex-lego_dataset,
title = { Hex: Lego Dataset },
type = { Open Source Dataset },
author = { HexHewwie },
howpublished = { \url{ https://universe.roboflow.com/hexhewwie/hex-lego } },
url = { https://universe.roboflow.com/hexhewwie/hex-lego },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2024 },
month = { jul },
note = { visited on 2024-12-03 },
}