Find cat at the image: visual recognition

How to detect what is on the uploaded picture? Cat or dog? If people, which age and gender? Here is a short manual.

Meet IBM Watson service! IBM Bluemix provides a lot of services like Amazon or Heroku. You just give a public URL to image to Visual Recognition service, and get back JSON with detailed information with recognized items, people or words. Find out more at the demo page visual-recognition-demo.mybluemix.net

How to do the same in your project? Registration and obtaining the key are implemented by very strange and un-user-friendly way, so I described it step-by-step.

1. Register at ibm.com/developerworks (make sure you set checkbox “Use Bluemix”) and confirm account.
2. Login to new-console.ng.bluemix.net with same credentials, do a lot of strange things what they will ask you.
3. Go to /catalog and find Visual Recognition service, step inside, read about pricing and click “Create” at the right bottom corner.
4. Wait for 5 minutes and get your credentials (API KEY).

If you are lost, you can find this information at /services.

Now you can go to Swagger docs watson-api-explorer.mybluemix.net/apis/visual-recognition-v3 and see the examples for your requests. It is enough to implement them throw simple HTTP GET.

Author: dobryakov

I’m a developer with full-stack business experience in web industry from sales to top-level technical management. Familiar with wide spectre of popular tools and methodics for whole life cycle of the software business - from communication with consumers to continuous delivery and support the long-term product. I have production experience in building the cloud and SaaS environments, designing distributed SOA systems and in DevOps activity. Feel free to contact me anytime by e-mail grigoriydobryakov@gmail.com