Free Information Center
BEVFormer (bird’s-eye-view transformer) is a deep learning architecture designed for 3D perception tasks, especially in autonomous driving, that integrates multi-camera inputs into a unified bird’s-eye-view representation using transformer models.
CARLA is an open-source autonomous driving simulator designed to support the development, training, and validation of autonomous vehicle systems. It provides realistic urban environments and sensor data for research and testing purposes.
The Deep Q-Network (DQN) is a pivotal reinforcement learning architecture applied in autonomous driving, enabling vehicles to make decisions based on visual inputs.
BEVDet is a deep learning framework designed for 3D object detection using bird’s-eye view representations. It is primarily used in autonomous driving systems to improve the perception of surrounding environments from multi-camera setups.
OccNet (occupancy network for driving) is a deep learning framework designed to model and predict the spatial occupancy of dynamic environments for autonomous driving. It uses 3D occupancy representations to enhance perception and decision-making in self-driving systems.
Imitation learning for autonomous driving is a machine learning approach where vehicles learn driving behaviors by mimicking human drivers. It aims to enable autonomous systems to replicate expert driving decisions through observation and data-driven modeling.
The A2D2 dataset is a comprehensive collection of data for autonomous driving, developed by Audi to aid in the research and development of self-driving technologies.