Menu

Neuronet

arrow

A serious challenge for E-2 was the request to develop mobile applications using artificial intelligence.

Our neural network improved the ability to recognize the smallest features of the face and hands practicing in areas such as physiognomy and chiromancy.

To train the neural network, we got a team of professional physiognomists and chiromantists. Dozens of thousands of hands and faces were processed.

As a result, we were able to reproduce not only the technical, but also the creative part of the process. At the same time we separated subjective distortions and bias of interpretation. After all, a physionomist or chiromantist can easily be succumbed to the temptation to distort the result a little in accordance with his opinion about the client.

Having incorporated the styles and methods of leading practitioners, impassive artificial intelligence has developed universal algorithms expressing the principles of chiromancy and physiognomy in the most pure form. Now we are working on the creation of two mobile applications where any user will be able to turn to our neural network for a physiognomic portrait and a hiromantic forecast.

FaceReaderProf is a physiognomy application.

It is based on a neural network that is able to recognize faces in great detail and interpret the data from the point of view of physiognomy.

The process of teaching such a neural network is knowledge-intensive and complex.

E2 group specialists took it through tens of thousands of images. As our faces differ from each other by millions of parameters, and the neural network should react more thinly to each of them.

The next step we taught the neural network to compare the data with the principles of physiognomy. The application has got a huge array of data, a system of knowledge from Aristotle to the leading experts of our time, allowing determining the character of personality by face.

And we did it! In the final report, FaceReaderProf characterizes each individual's personality based on its characteristics of face features.

For example, wide-eyed people speak of tolerance, condescension, both to themselves and to others.

Testing the application has shown that in most cases the system works without errors. In addition, artificial intelligence is biased and free from emotions. It describes not subjective impressions, but gives physiognomy in the most pure form.

Now, just by uploading a photo to our neural network, everyone will be able to get a physiognomic personality analysis based on science.

Mobile application on chiromancy

The next step on our neuropath was to create a application on chiromancy. PalmReaderProf.

The idea was to teach the neural network to recognize the human hand very accurately.

The most interesting thing was to come up with exactly how to teach it. As no one has any experience in palm recognition.

It was important for us to achieve very high quality recognition.

Identification of each of its elements, palms, phalanx on fingers, joints.

And we have done it! It turned out to be a very powerful model of the neural network, which determines the elements of the hands with 98.7 percent accuracy. It mathematically calculates how much wider, shorter or longer these elements are as relating to each other.

The next step is to put an array of knowledge on chiromancy into the application, including the experience of leading experts in the field.

And as a result, we got an application that can do psychological analysis of the personality only by a photo of the palm. At the same time, it is more efficient and reliable than the most experienced specialist in this field.

Apps that read people's faces and hands are created as part of the development of artificial intelligence, capable of interacting with people and solving creative problems. The commercial version of the app is scheduled for launch in the summer of 2019.

Projects

Contacts

+7(4212) 65-15-20
manager@e2.group

501 office, 5th floor, 132 Kalinina str, Khabarovsk, Russia

Sent

Your message has been sent,
we will contact you shortly.

Feedback