In this repo we store our collected images from local supermarkets Magnit and Pyaterochka. Also, we train and test different YOLO models on grocery object detection task.
Custom Grocery dataset (items from local Russian supermarkets)
Targeted labels (as the most common bought items):
- Bread
- Milk
- Eggs
- Tomatoes
- Carrot
- Cucumber
- Apples
- Bananas
- Meat
- Sausages
- Cheese
Follow our telegram bot's reporistory. Check the instructions there. In short, you need to run docker compose file with <docker-compose up> command
