Conditioning Latent-Space Clusters for Real-World Anomaly Classification

Abstract

Anomalies in the domain of autonomous driving are a major hindrance to the large-scale deployment of autonomous vehicles. In this work, we focus on high-resolution camera data from urban scenes that include anomalies of various types and sizes. Based on a Variational Autoencoder, we condition its latent space to classify samples as either normal data or anomalies. In order to emphasize especially small anomalies, we perform experiments where we provide the VAE with a discrepancy map as an additional input, evaluating its impact on the detection performance. Our method separates normal data and anomalies into isolated clusters while still reconstructing high-quality images, leading to meaningful latent representations.

Daniel Bogdoll, Svetlana Pavlitska, and Simon Klaus contributed equally. Accepted for publication at SSCI 2023

References in corpus (7)

• Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
• Anomaly Detection with Adversarial Dual Autoencoders
• Perception Datasets for Anomaly Detection in Autonomous Driving: A Survey
• Ad-datasets: a meta-collection of data sets for autonomous driving
• Corner Cases for Visual Perception in Automated Driving: Some Guidance on Detection Approaches
• Latent space conditioning for improved classification and anomaly detection
• Impact, Attention, Influence: Early Assessment of Autonomous Driving Datasets