… The first 12 models all use the Faster R-CNN meta-architecture. In Table II, We use car,
ped and all to respectively represent three detection tasks which include the detection of cars, …

… autonomous driving. Therefore, cloud-assisted AV (CAV) [4], [6], [7] was introduced to support
autonomous driving… from the AV manufacturers and autonomous driving service providers. …

… We believe this is important for further advances in autonomous driving, and therefore have
… We also include vehicle bus data, which provides additional information about car state (eg …

… approach to automated scenario-based testing of the safety of autonomous vehicles, especially
… Experiments with a real autonomous vehicle at an industrial testing facility support our …

… Autonomous driving is developing rapidly and has … in autonomous vehicles, to the best of
our knowledge, none has investigated the security of lane detection module in real vehicles. In …

… Abstract—Multi-object tracking is an important ability for an autonomous vehicle to safely
navigate a traffic scene. Current state-of-the-art follows the tracking-by-detection paradigm …

… We present InterFuser, a novel autonomous driving framework based on an interpretable
sensor fusion transformer. The driving agent obtains a perception of global context from multi-…

… commonly used sensors in autonomous driving: cameras and … from multiple views, we first
use an image backbone (such as … in end-to-end autonomous driving which are too sparse for …

… autonomous driving. Monocular 3D detection is a challenging task because of the lack of
accurate 3D information in a single image. However, the huge potential in such a cheap and …

… driving research based on Unreal Engine 4. … settings, and multiple vehicles with different
physical parameters. In our experiments we use input from simulated cameras; their settings, …