Identifying Out-of-Distribution Samples in Real-Time for Safety-Critical 2D Object Detection with Margin Entropy Loss

no code implementations • 1 Sep 2022 • Yannik Blei, Nicolas Jourdan, Nils Gählert

Convolutional Neural Networks (CNNs) are nowadays often employed in vision-based perception stacks for safetycritical applications such as autonomous driving or Unmanned Aerial Vehicles (UAVs).

Autonomous Driving Object +3