Google Cloud Vision API Client: Flask + tfPhotoPalette
Switch branches/tags
Nothing to show
Clone or download
Latest commit 50cb2ba Mar 26, 2016
Permalink
Failed to load latest commit information.
.gitignore add ignore Mar 26, 2016
.python-version add ignore Mar 26, 2016
LICENSE add MIT License Mar 26, 2016
README.md on Mar 26, 2016
palette_server.py initial commit Mar 26, 2016
pip_modules.txt initial commit Mar 26, 2016

README.md

Google Cloud Vision API Client

This is a GUI tool to try the feature LABEL_DETECTION of Google Cloud Vision API. https://gyazo.com/0aa5cddd30b8bbc956aa6ced95fd244c

Set your API key

$ echo "CLOUD_VISION='YOUR_API_SERVER_KEY'\n" >> googleapikeys.py

Install pip modules

$ pip install -r pip_modules.txt

Install tfPhotoPalette

You can get it in Chrome Web Store: https://chrome.google.com/webstore/detail/tfphotopalette/gcpfanfkkjpolcdicokfjphmdnelhbbb

Run local server

$ python palette_server.py

Run tfPhotoPalette

You can launch tfPhotoPalette app in following steps:

  • Open chrome://apps/
  • Click the app icon of tfPhotoPalette

Send requests

Give your favorite photo. You can give a photo in following two ways.

  • Drag and drop a photo file onto tfPhotoPalette app.
  • Enter the URL of a photo on the Web and click "Load" button.

https://gyazo.com/1484dd1d17d3261f550b695f2e8d3a2f

And then, click "Play" button in right-bottom of the window. The small window for sending request to local server will open. Select the value Original photo in a left-top select button. In the last step, click "Classify" button to send LABEL_DETECTION request to Google Cloud Vision API through the local server you launched earlier.

Get response

You can get labelAnnotations results. https://gyazo.com/14f57b3eee0b2f636251a5e9f089984d

Please go to http://daiiz.hatenablog.com/entry/2016/03/26/020347 for more details.