Google Cloud Vision API is a powerful, off the shelf tool that provides you with Image Recognition, Object Detection, and OCR. There is also a RoogleVision package that makes it really easy to use the service.
Answers to the exercises are available here.
If you obtained a different (correct) answer than those listed on the solutions page, please feel free to post your answer as a comment on that page.
Exercise 1
Install RoogleVision. Sign in at Google Cloud Platform. Create a project, enable billing (free up to 1000 requests per month) and enable Google Cloud Vision API. To be able to connect with the Google Cloud, create OAuth 2.0 client ID at the credentials tab. Finally, connect with the Google Cloud.
Exercise 2
Detect what is on the image below.
Exercise 3
Detect the landmark on the image from exercise 2.
Exercise 4
Plot the image from exercise 2 with a box around the detected landmark. (You will need an additional package, for example EBImage, for this one.)
Exercise 5
Mark on the map the location of the landmark. (You will need an additional package, for example leaflet, for this one.)
- Work extensively with the GoogleVis package and its functionality
- Learn what visualizations exist for your specific use case
- And much more
Exercise 6
Detect a logo on the picture below and mark it with a box.
Exercise 7
Detect a face on the picture below and mark it with a box.
Exercise 8
Mark face landmarks on the image.
Exercise 9
What is the most probable emotion expressed by the face on the image?
Exercise 10
Get the text from the image below.
Leave a Reply