Neural Network Meta Classifier: Improving the Reliability of Anomaly Segmentation

TL;DR

This paper enhances anomaly segmentation in neural networks by replacing logistic regression with a lightweight neural network meta classifier, improving detection reliability in open-set environments like road driving.

Contribution

It introduces a neural network meta classifier for anomaly detection, demonstrating its advantages over logistic regression and analyzing the trade-offs involved.

Findings

Neural network meta classifier outperforms logistic regression in anomaly detection.

Incorporating informative out-of-distribution examples improves training results.

Behavior of logistic regression and neural network classifiers is strongly correlated.

Abstract

Deep neural networks (DNNs) are a contemporary solution for semantic segmentation and are usually trained to operate on a predefined closed set of classes. In open-set environments, it is possible to encounter semantically unknown objects or anomalies. Road driving is an example of such an environment in which, from a safety standpoint, it is important to ensure that a DNN indicates it is operating outside of its learned semantic domain. One possible approach to anomaly segmentation is entropy maximization, which is paired with a logistic regression based post-processing step called meta classification, which is in turn used to improve the reliability of detection of anomalous pixels. We propose to substitute the logistic regression meta classifier with a more expressive lightweight fully connected neural network. We analyze advantages and drawbacks of the proposed neural network meta classifier and demonstrate its better performance over logistic regression. We also introduce the concept of informative out-of-distribution examples which we show to improve training results when using entropy maximization in practice. Finally, we discuss the loss of interpretability and show that the behavior of logistic regression and neural network is strongly correlated.