Elbit Systems will discuss real time semantic segmentation using GPUs both on manned and unmanned aerial vehicles, and its applications in landing optimization. A large part of this work consists of designing a novel fully connected network that can handle 3D data as well as various 2D inputs (multi-spectral), 3D classification, developing methods for real time classification on point clouds using deep CNN. In the field of AI, they will discuss reinforcement and imitation learning as part of autonomous entities, anomaly detection on large scale databases (using DNNs), and time series and predictions on sensor data to identify loads on pilots.
Hall: Hall F
Track: Cyber Security