Researchers have trained a machine learning algorithm to create ground-level images for the pictures taken by the satellite. They used a machine learning system known as cGAN which combines two neural networks: a ‘generator’ and a ‘discriminator’.
An artificial intelligence model managed to create a script, facial expressions, voice and select scenes to make an end-to-end movie. The neural network was built using TensorFlow, GANs and AWS. Check out the video inside!