LS-VOS: Identifying Outliers in 3D Object Detections Using Latent Space Virtual Outlier Synthesis

TL;DR

LS-VOS enhances 3D object detection in autonomous driving by effectively identifying outliers through latent space synthesis, improving safety and reliability without sacrificing detection accuracy.

Contribution

The paper introduces a novel latent space-based outlier synthesis method that improves outlier detection in 3D object detectors.

Findings

Improved outlier detection performance in 3D detection models

Maintained high detection accuracy while identifying outliers

Effective outlier synthesis using auto-encoder latent space

Abstract

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.