Browse » Favorite
Top Favorite Datasets
Open source favorite computer vision datasets, pre-trained models, and APIs.
Overview
The GDIT Aerial Airport dataset consists of aerial images containing instances of parked airplanes. All plane types have been grouped into a single classification named "airplane".
Example Image
Background Information
This dataset was created as part of the World's Largest Game of Rock, Paper, Scissors talk and challenge introduced by Joseph Nelson and Salo Levy @ SXSW 2023.
- the above image is linked to the entry page
- the demo video was prepared from the Deploy Tab, and utilizes
v11
(YOLOv8n-100epochs
)
The dataset includes an aggregation of images cloned from the following datasets:
- https://universe.roboflow.com/brad-dwyer/egohands-public/ - null images
- https://universe.roboflow.com/presentations/rock-paper-scissors-presentation/
- https://universe.roboflow.com/team-roboflow/rock-paper-scissors-detection
- universe.roboflow.com/popular-benchmarks/mit-indoor-scene-recognition/
New images were added to the dataset and labeled to supplement the examples from the cloned datasets. Members of Team Roboflow, and more close friends of the team, are included in the dataset to assist with creating a more robust, generalized, model.
Participation Rules and FAQ
- the above image is linked to the FAQ and contest entry page
Logistics Pre-trained Object Detection Model
Pre-trained models are trained on large datasets until they achieve good generalization, meaning they can recognize patterns effectively. "pre-trained" indicates that the model has already undergone training on a substantial dataset, often a generic one, and is ready for fine-tuning on a specific task with a smaller dataset. The Logistics Object Detection Base Model is a pre-trained model hosted on Roboflow Universe, created to be a strong starting point for custom training on logistics-specific object detection tasks. This model is built on a dataset of 99,238 images across 20 logistics-focused classes, collected from various projects on Roboflow Universe. Part of this dataset was auto-labeled using the Autodistill DETIC tool from Roboflow, helping to achieve a mean Average Precision (mAP) of 76%.
Classes:
- Barcode, QR Code
- Car, Truck, Van
- Cardboard Box, Wood Pallet, Freight Container
- Fire, Smoke
- Forklift
- Gloves, Helmet, Safety Vest
- Ladder
- License Plate
- Person
- Road Sign, Traffic Cone, Traffic Light
Current Status: The model has achieved a mAP of 76%, marking its readiness as a checkpoint for further custom training. It aims to shorten the development cycle, facilitating better model performance in specific logistics scenarios.