Unlimited Resolution Image Generation with R2D2-GANs
This provides a solution for generating synthetic sonar data to train autonomous recognition systems, particularly in domains like seabed mapping, though it is incremental as it builds on existing GAN methods.
The paper tackles the problem of generating high-resolution sonar images by introducing a conditional recursive GAN architecture, enabling unlimited resolution image synthesis that is indistinguishable from real scans and can be generated at least twice as fast as real acquisition.
In this paper we present a novel simulation technique for generating high quality images of any predefined resolution. This method can be used to synthesize sonar scans of size equivalent to those collected during a full-length mission, with across track resolutions of any chosen magnitude. In essence, our model extends Generative Adversarial Networks (GANs) based architecture into a conditional recursive setting, that facilitates the continuity of the generated images. The data produced is continuous, realistically-looking, and can also be generated at least two times faster than the real speed of acquisition for the sonars with higher resolutions, such as EdgeTech. The seabed topography can be fully controlled by the user. The visual assessment tests demonstrate that humans cannot distinguish the simulated images from real. Moreover, experimental results suggest that in the absence of real data the autonomous recognition systems can benefit greatly from training with the synthetic data, produced by the R2D2-GANs.