1

I am familiar with watson API key visual recognition :

visual_recognition = VisualRecognitionV3('2016-05-20', api_key=param)

with param as my api key.

this Api Call is nice to recognize forms. But Is it possible to get the positions of the elements that are recognized ? I do not see it in the API documentation. I see this is possible for face detection, but let say, I have a dog on my picture, how can I have the postions of this dogs ?

Here is the API reference

Sayuri Mizuguchi
  • 5,250
  • 3
  • 26
  • 53
MouIdri
  • 1,300
  • 1
  • 18
  • 37

1 Answers1

2

The service works by using a collection of classifiers, each classifier is a single tag only and must be trained with it's own sets of positive and negative images. Try to create smaller sub-images out of one image containing multiple entities.

See one video to verify how it works.

Fork on Github.

Sayuri Mizuguchi
  • 5,250
  • 3
  • 26
  • 53
  • I will upvote your response and mark my question as answered. However, I would like to have the Python way to do this ( not the node ). – MouIdri May 23 '17 at 13:01
  • 1
    Here have some example to use with [Python](https://www.ibm.com/blogs/watson/2016/09/power-visual-recognition-use-watson-identify-hand-cards/). I show one example with Node because have how to do what you want, the logic is the same, you'll just did the same with other programming language – Sayuri Mizuguchi May 23 '17 at 13:17
  • Thanks for this. I already do some advanced usage of visual rec and python I really needed the localisation of the position using python language. I will have have a look. – MouIdri May 23 '17 at 13:34