An Outlier Exposure Approach to Improve Visual Anomaly Detection Performance for Mobile Robots
This work addresses anomaly detection for mobile robots, but it is incremental as it builds on existing Real-NVP models with an auxiliary loss.
The paper tackles the problem of improving visual anomaly detection for mobile robots by leveraging available anomaly examples, and shows that their approach outperforms alternatives and yields significant performance improvements on a novel indoor patrolling dataset.
We consider the problem of building visual anomaly detection systems for mobile robots. Standard anomaly detection models are trained using large datasets composed only of non-anomalous data. However, in robotics applications, it is often the case that (potentially very few) examples of anomalies are available. We tackle the problem of exploiting these data to improve the performance of a Real-NVP anomaly detection model, by minimizing, jointly with the Real-NVP loss, an auxiliary outlier exposure margin loss. We perform quantitative experiments on a novel dataset (which we publish as supplementary material) designed for anomaly detection in an indoor patrolling scenario. On a disjoint test set, our approach outperforms alternatives and shows that exposing even a small number of anomalous frames yields significant performance improvements.