2017/05/27 by Edgar Sucar, Sucar, Edgar, Jean-Bernard Hayet +1
Computer Science · Engineering · Environmental Science · #Advanced Vision and Imaging #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Remote Sensing and LiDAR Applications #Robotics and Sensor-Based Localization
paper · pdf · doi:10.48550/arxiv.1705.09860
openalex publication_date 2017/05/27 · openalex created_date 2022/08/19 · openalex updated_date 2026/07/28
This paper proposes a novel method to estimate the global scale of a 3D\nreconstructed model within a Kalman filtering-based monocular SLAM algorithm.\nOur Bayesian framework integrates height priors over the detected objects\nbelonging to a set of broad predefined classes, based on recent advances in\nfast generic object detection. Each observation is produced on single frames,\nso that we do not need a data association process along video frames. This is\nbecause we associate the height priors with the image region sizes at image\nplaces where map features projections fall within the object detection regions.\nWe present very promising results of this approach obtained on several\nexperiments with different object classes.\n