This is an image tagger that can tag images from all around the web.
The model is trained from images from Instagram, Google and OpenImages.
- To begin, place images inside the data folder. Images of each category must be placed inside its respective folder (category = folder name)
- Next run the final model, which will train an image tagger and store the final result.
The model comes with a web interface that can tag images on the fly.
- Run the backend server (
python img/server.py) - RUn the frontend
cd img/image-taggerng serve