Appearance-Invariant 6-DoF Visual Localization using Generative Adversarial Networks
This work tackles the problem of robust long-term visual localization for autonomous systems operating in dynamic outdoor environments, providing an incremental improvement over existing methods.
This paper addresses visual localization under varying environmental conditions by proposing a network that extracts appearance-invariant features using a CycleGAN-based encoder. This approach enables robust 6-DoF pose regression despite changes in illumination, weather, and season, outperforming state-of-the-art methods on challenging datasets.
We propose a novel visual localization network when outside environment has changed such as different illumination, weather and season. The visual localization network is composed of a feature extraction network and pose regression network. The feature extraction network is made up of an encoder network based on the Generative Adversarial Network CycleGAN, which can capture intrinsic appearance-invariant feature maps from unpaired samples of different weathers and seasons. With such an invariant feature, we use a 6-DoF pose regression network to tackle long-term visual localization in the presence of outdoor illumination, weather and season changes. A variety of challenging datasets for place recognition and localization are used to prove our visual localization network, and the results show that our method outperforms state-of-the-art methods in the scenarios with various environment changes.