Overview of My Submission
Manohari is a web application that tells how much attractive a person is for others, it works on some studies that I will tell you later in this blog.
It is built with Azure App Service, Azure Blob Storage and Azure Cognitive Services (Face API).
Go give it a try at https://manohari.azurewebsites.net and check yourself.
Submission Category:
Wacky Wildcards
Link to Code on GitHub
Last weekend, I asked my friend about a silly game that we played in childhood and can build for hackathon. He told me a love calculator and I took it seriously.
Tell me in the comments, how much you like this silly idea.
Manohari
Find your attractive score using this application, it works on a simple principle of similar facial characteristics between persons. Give it a try at https://manohari.azurewebsites.net, My Project Submission to Microsoft Azure Trial Hackathon on DEV.
It is built using Azure App Service, Azure Blob Storage and Azure Cognitive Service (Face API).
Working
According to some studies it is found that people likes faces that have common facial characteristics to them or their near one. So, I used Azure Face API to match faces and get the confidence, after that I tried to neutralise gender biasing and added emotional score. I added some quotes to tell your some points that I consider while calculating the score.
It is not much accurate because of two reasons:
- It differs from person to person and depends upon person perspective.
- Azure Face API tries to match person and doesn't returns the score for similar facial characteristics.
I believe that if I train my own ML model then it will produce more accurate results but it is only for fun. I handed the case if same person provided in both images, it checks that the persons are identical and returns no score.
You can understand my algorithm from my GitHub repository provided and tell me in the comments how I can improve it.
Azure Services
I used three Azure Services:
- Azure App Service
- Azure Face API
- Azure Blob Storage
Flask application hosted on Azure App Service provides user interaction and communicate with Azure Blob Storage and Face API. Blob storage is used to storage the images provided by user and Face API uses those images to detect and find similar.
Screenshots and Gif
Demo
Angry Face
Disgusting Face
End Note
I hope you find this interesting and tell me in the comments about your opinion on it.
Thanks to my friend Chetan Goyal to provide this idea.
Top comments (0)