Ensemble of Experts for Generating Digital Parking Maps from Remote Sensing Images

Being a vital component of transportation, parking and parking-related research are gaining increasing attention from the intelligent transportation and automotive industry research communities. A digital parking map with precise parking spot geospatial information is crucial for research such as automatic valet parking, parking spot recommendations, and parking route optimization. Currently, these maps are generated through costly and resource-intensive human annotation. To address the challenges of insufficient parking maps and the high costs of manual map creation, researchers have proposed the following two strategies for generating parking maps. The SLAM-based map generation for indoor parking lots and the remote sensing image-based map generation for outdoor parking lot. Our work mainly focuses on the outdoor high-definition parking map generation from remote sensing images. These images often suffer from occlusion, inconsistent resolution, and varying luminosity conditions. To address these limitations, in this paper, we propose an ensemble-of-experts (EoE) based parking map generation scheme that robustly produces vectorized parking maps, extracting the location and geometry of parking spots from remote sensing images. The proposed scheme comprises a EoE module and a vector map creation module. The EoE module integrates four machine learning models focused on vehicle detection, parking lot segmentation, parking block segmentation, and tree occlusion detection and segmentation. These four expert models are designed to address various edge cases encountered in parking map generation tasks using remote sensing images.The experiments demonstrate that the proposed scheme achieves an 80.5% parking spot detection precision and a center-to-center geometric representation error of 0.93 meters for detected parking spots.