Microsoft Cognitive Services – Connect the Dots with AI Services

What are Cognitive Services?

Microsoft Cognitive Services is a set of cloud services available for developers to build intelligent applications without having direct Machine Learning or Data Science knowledge. Services are ready to be consumed through easy-to-use APIs without any hassle. Microsoft highlights this catalog of services within the Azure stack as another supportive initiative towards democratization of artificial intelligence. The idea behind this is that every person and every organization should be able to access the AI capabilities.

This comprehensive library of services is becoming a game changer in many competitive business domains. Sabre, Prism Skylabs and Uber are not only few of the businesses already enjoying these services, but has been highlighted as successful in accomplishing their intended outcomes.

These are the main categories of available services;

Why Cognitive Services

• Easy to use – It’s a matter of accessing APIs through REST calls. In fact, it’s easy to integrate with any application
• Interoperability – Services can be utilized in any language, framework or platform. Services cater to a wide audience.
• Well tested – Each API is a result of Microsoft Research and has gone through an extensive testing procedure before being made available.
• Growing – Different services that address new challenges are being released regularly.
• Cost effectiveness – Less costly since they are pay-as-you-go services. Also, it’s worth considering its advantages, such as credibility of service provider, being a tested solution and having a high-standard infrastructure which services are hosted on.

Closer look at Vision API

Vision API is an exciting API which is getting popular day-by-day. It recognizes different objects and `It should be noted that, accuracy and level of granularity in detail would grow with time, enabling more real-world applications.

The above illustration shows how an image is analyzed. Yet, Vision API can be used in many other ways such as to read text in an image, to read handwritten text, to identify a celebrity, etc.

Just as Vision API, other APIs such as Speech, Search, etc. work well to address relevant scenarios with decent accuracy.

Uber’s Real-Time ID Check

One effective solution given to solve a real business problem through Microsoft Cognitive Services (Vision API to be exact) is Uber’s Real-Time ID Check.

An Uber driver is requested to take a selfie once a service request is accepted in order to verify that the driver is the same person who underwent Uber’s screening process.

Below are the main steps of the facial verification process;

• 1. Face Detection: Identify a face in a given image. If a face is not detected, the next steps are skipped as there is no face to match.


• 2. Feature Extraction: Build a feature set by extracting facial features, like eye position and face shape from the image.


• 3. Face Matching: Compare the feature set with an already validated feature set for a given face.

The business has selected Microsoft Face API of Cognitive Services because it was the most accurate service and it produced the strongest set of critical features demanded by the business case. In order to cater to the above process, it utilizes two separate functionalities as mentioned below;

Below are the main steps of the facial verification process;

• 1. Face Detect: Detects human faces in an image and provides additional attributes, such as whether the face has glasses. This API is smart enough to also differentiate between reading glasses and sunglasses. This helped Uber filter out images without a face and improved the match score for verification of the selfies.


• 2. Face Verify: Compares a face detected from Face-Detect API with an already verified face and provides a confidence score on whether the two faces belong to the same person or not. Based on the confidence score, Uber can take appropriate verification action, such as asking the user to retake the selfie.

So, this is one creative solution provided, utilizing Microsoft Cognitive services to solve a real business problem. More such exciting applications would propagate in the near future. While tools are readily available, what matters in this moment is how creative you are in connecting the dots.

References

www.docs.microsoft.com
www.theflyingmaverick.com
www.blogs.windows.com
www.eng.uber.com
www.news.microsoft.com