This repo is a demo using google vision api to analyze images.
The input should be a csv file listing all images and a folder containing images to be analyzed:
The output should be an updated csv file with a new column added - "labels":
Taking one of the sample images as an example, for this image:
the detected labels are:['furniture', 'dining room', 'table', 'property', 'room', 'chair', 'interior design', 'floor', 'flooring', 'kitchen & dining room table']
Please make sure you have a google cloud account and set up a payment method. You can set it up here: https://cloud.google.com
After register, log in to your console:https://console.cloud.google.com, create a project and enable your api:https://cloud.google.com/vision/docs/before-you-begin
Because we will call google api locally, first we need to install google cloud sdk.
Enter the following at terminal: curl https://sdk.cloud.google.com | bash
Restart terminal and enter: exec -l $SHELL
Run gcloud init to initialize the gcloud environment:gcloud init You will need to log in your google account. Choose the project you just created.
Install Auth components: gcloud beta auth application-default login and create a service account: https://developers.google.com/identity/protocols/OAuth2ServiceAccount#creatinganaccount After this you will have a .json file downloaded, for example I save it as ~/visiontesting.json
Set environment variables: export GOOGLE_APPLICATION_CREDENTIALS=~/visiontesting.json
Google cloud sdk uses Python 2.7, so this demo will also use Python 2.7. Make sure that Python 2.7 is installed on your system:python -V
If you have not installed Python yet, please install Python 2.7:https://www.python.org/downloads/
-
Create virtual environment:
$virtualenv venv -
Activate virtual env:
$source venv/bin/activate -
install required packages:
pip install -r requirements.txt -
Run demo:
python label_add.py
If you meet with error "No module named 'google.cloud'", try to use: pip install --upgrade google-cloud and it should work.