Occupancy grid mapping is an important component in road scene understanding for autonomous driving. It encapsulates information of the drivable area, road obstacles and enables safe autonomous driving. Radars are an emerging sensor in autonomous vehicle vision, becoming more widely used due to their long range sensing, low cost, and robustness to severe weather conditions. Despite recent advances in deep learning technology, occupancy grid mapping from radar data is still mostly done using classical filtering approaches. In this work, we propose learning the inverse sensor model used for occupancy grid mapping from clustered radar data. This is done in a data driven approach that leverages computer vision techniques. This task is very challenging due to data sparsity and noise characteristics of the radar sensor. The problem is formulated as a semantic segmentation task and we show how it can be learned using lidar data for generating ground truth. We show both qualitatively and quantitatively that our learned occupancy net outperforms classic methods by a large margin using the recently released NuScenes real-world driving data.

No abstract

No abstract

No abstract

BIM is a multi-template matching algorithm. As opposed to traditional template matching algorithms that match a single template to a single image, BIM attempts to match multiple templates to a single image at once. A naive approach to multi-template matching would be to run a standard template matching algorithm sequentially with each of the templates and report the best result. Instead, each template processed by BIM further restricts the search space of the following templates, thus speeding up the overall process. In particular, we extend a recently introduced method for single template matching under 2D affine transformation to work with multiple templates at once. As a result, given a library of templates we can efficiently find the best 2D affine transformation for each of them in a target image. Experiments on real data sets reveal speedups of between 10 and 17.