Abstract
Long-Term visual localization under changing environments is a challenging problem in autonomous driving and mobile robotics due to season, illumination variance, etc. Image retrieval for localization is an efficient and effective solution to the problem. In this paper, we propose a novel multi-task architecture to fuse the geometric and semantic information into the multi-scale latent embedding representation for visual place recognition. To use the high-quality ground truths without any human effort, depth and segmentation generator model is trained on virtual synthetic dataset and domain adaptation is adopted from synthetic to real-world dataset. The multi-scale model presents the strong generalization ability on real-world KITTI dataset though trained on the virtual KITTI 2 dataset. The proposed approach is validated on the Extended CMU-Seasons dataset through a series of crucial comparison experiments, where our performance outperforms state-of-the-art baselines for retrieval-based localization under the challenging environment.
Abstract (translated)
URL
https://arxiv.org/abs/2010.00573
