Some of them are unable to retain the actual identity of the faces during the process, or it might be very expensive and time-consuming. However, you will feel good to know that researchers have just used machine learning to design a system that can make you look 30 or 40 years older with 80% accuracy. Grigory Antipov from France’s Orange Lab has developed the deep learning machine with the help of his teammates. Guess what? The computer system makes the faces to look older and younger. The technique is named Age Conditional Generative Adversarial Network. According to the reports from Technology Review, the approach used by Grigory Antipov and his team involves two deep-learning machines that work together. One for Face Generator and another one is for Face Discriminator. These machines are then trained using 5,000 faces that were picked from IMDB and Wikipedia in the age groups 0-18, 19-29, 30-39, 40-49, 50-59, and 60+ years old. The machine was able to learn the signature feature of faces in each age group that is used to apply on other faces. However, using the signature on different faces can sometimes cause a person’s identity to be lost. In that case, the second deep-learning machine – The Face Discriminator studies the synthetically generated faces to see either the real identity can still be pulled out. If it fails to do so, then the image is rejected. The researchers claimed, that their system spotted the old faces correctly about 80% of the time, compared to about 50% of the time for other techniques which relates to face aging. Well, you all now might be thinking what the use of such system is? Researchers claimed that their techniques could be used to find people who go missing for many years. So, what do you think about this? Share your views in the comment box below!

Δ

This Computer System Displays How Your Face Will Look Like After 40 Years - 36This Computer System Displays How Your Face Will Look Like After 40 Years - 45This Computer System Displays How Your Face Will Look Like After 40 Years - 52