LS-VOS: Identifying Outliers in 3D Object Detections Using Latent Space Virtual Outlier Synthesis
This addresses safety and reliability issues in autonomous vehicles by reducing false detections, though it is an incremental improvement on existing outlier synthesis methods.
The paper tackles the problem of unreliable 3D object detections in autonomous driving by proposing LS-VOS, a framework that improves outlier detection capabilities while maintaining high detection performance, as demonstrated in extensive experiments.
LiDAR-based 3D object detectors have achieved unprecedented speed and accuracy in autonomous driving applications. However, similar to other neural networks, they are often biased toward high-confidence predictions or return detections where no real object is present. These types of detections can lead to a less reliable environment perception, severely affecting the functionality and safety of autonomous vehicles. We address this problem by proposing LS-VOS, a framework for identifying outliers in 3D object detections. Our approach builds on the idea of Virtual Outlier Synthesis (VOS), which incorporates outlier knowledge during training, enabling the model to learn more compact decision boundaries. In particular, we propose a new synthesis approach that relies on the latent space of an auto-encoder network to generate outlier features with a parametrizable degree of similarity to in-distribution features. In extensive experiments, we show that our approach improves the outlier detection capabilities of a state-of-the-art object detector while maintaining high 3D object detection performance.