Installation

1. Clone this repository

https://github.com/MasumBhai/Dataset-preparation-for-YOLO.git

2. Requirement installation

pip3 install -r requirements.txt

Where to get labeled image dataset

Open Image Dataset

How to download those labeled image dataset

For Single Class Download to train & validation

python main.py downloader --classes "Missile" --type_csv train --limit 300
python main.py downloader --classes "Missile" --type_csv validation --limit 300

For Multiple Class Download to train & validation

python main.py downloader --classes "Ball" "Football" "Golf_ball" "Cricket_ball" "Tennis_ball" --type_csv train --limit 500
python main.py downloader --classes "Ball" "Football" "Golf_ball" "Cricket_ball" "Tennis_ball" --type_csv validation --limit 300

How to Annotate those downloaded images

python convert_annotations.py

Now, just remove label folder from each classes inside OID -> Dataset -> train & validation folder for YOLO processing

Acknowledgement

these two awesome repositories

1. theAIGuysCode / OIDv4_ToolKit

2. EscVM / OIDv4_ToolKit

Feedback

If you have any feedback, please reach out to me at [email protected]