Quantcast
Channel: NewsCrunch
Viewing all articles
Browse latest Browse all 2337

Microsoft’s emotion recognition engine goes live – upload your photo and check what your mood is

$
0
0
Microsoft has released a new tool that will analyse facial expressions and identify emotions which are conveyed by them.  Just go to its Emotion Recognition page and upload a photo. The emotion tool will analyse and assign values to eight core emotional states—anger, contempt, fear, disgust, happiness, neutral, sadness or surprise.
Young Amitabh Bachchan is both angry and neutral,
according to Microsoft’s emotion recognition engine


 November 13, 2014, NewsCrunch

Microsoft has released a new tool that will analyse facial expressions and identify emotions, which are conveyed by them.

you can go to its Emotion Recognition page and upload a photo. The emotion tool will analyse and assign values to eight core emotional states—anger, contempt, fear, disgust, happiness, neutral, sadness or surprise.

Microsoft has developed this tool not just to amuse you. It is part of Project Oxford, which is working to provide developers with machine learning tools that’ll help them build personalised apps.

So, how good is it? It is in beta stage and not very accurate. It assigns nearly similar angerand neutral values to the above photo of young Amitabh Bachchan, the famous angry face of India of the 70s.

So, how good is it? It is in beta stage and not very accurate. It seems to think the above photo of young Amitabh Bachchan, the famous angry face of India of the 70s, is both angry and happy at the same time.   

Here are the values assigned for Amitabh Bachchan’s angry young man photo by Microsoft’s emotion recognition engine



Anger: 0.0776313543,
Contempt: 0.118609972,
Disgust": 0.07158525,
Fear: 0.0007974382,
Happiness": 0.0128688095,
Neutral": 0.702376366,
Sadness": 0.0123327859,

Surprise": 0.00379801355

Viewing all articles
Browse latest Browse all 2337

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>