Addressing Data Annotation Challenges in Multiple Sensors: A Solution for Scania Collected Datasets

arXiv:2403.18649v1 Announce Type: new
Abstract: Data annotation in autonomous vehicles is a critical step in the development of Deep Neural Network (DNN) based models or the performance evaluation of the perception system. This often takes the form of adding 3D bounding boxes on time-sequential and registered series of point-sets captured from active sensors like Light Detection and Ranging (LiDAR) and Radio Detection and Ranging (RADAR). When annotating multiple active sensors, there is a need to motion compensate and translate the …

abstract annotation arxiv autonomous autonomous vehicles challenges cs.cv cs.sy data data annotation datasets deep neural network development dnn eess.sy evaluation form multiple network neural network perception performance sensors series solution type vehicles