Enhancing Detail of 3D Terrain Models Using GAN

摘要

The paper addresses the problem of low quality 3D terrain models enhancement. We propose the approach based onconvolutional neural networks (CNN), namely, on Pix2Pix method that uses generative adversarial networks for imageto-image translation. We use heightmap 3D terrain models representation to use classical CNNs. The network was trainedon a synthetic dataset that included 150000 images and heightmaps of different landscapes. Our model showed the relativemean absolute difference equal to 0.459% on synthetic testing dataset. In addition, we demonstrate landscapes generationon the real data from Google Maps using our model.