Closing the Reality Gap with Unsupervised Sim-to-Real Image Translation

TL;DR

This paper presents an unsupervised image translation method that enhances synthetic images with real-world styles, significantly improving real-time neural network performance on autonomous robots by bridging the simulation-to-reality gap.

Contribution

Introduces a novel unsupervised style transfer technique for synthetic images, improving the realism of simulation data for training robotic vision systems.

Findings

Enhanced model accuracy with style-transferred images

Significant performance boost over purely simulated data

Effective real-time application on autonomous robots

Abstract

Deep learning approaches have become the standard solution to many problems in computer vision and robotics, but obtaining sufficient training data in high enough quality is challenging, as human labor is error prone, time consuming, and expensive. Solutions based on simulation have become more popular in recent years, but the gap between simulation and reality is still a major issue. In this paper, we introduce a novel method for augmenting synthetic image data through unsupervised image-to-image translation by applying the style of real world images to simulated images with open source frameworks. The generated dataset is combined with conventional augmentation methods and is then applied to a neural network model running in real-time on autonomous soccer robots. Our evaluation shows a significant improvement compared to models trained on images generated entirely in simulation.