Improving traffic sign recognition by active search

TL;DR

This paper presents an active-learning method that iteratively improves traffic sign recognition by efficiently identifying rare signs from unlabeled data, even starting from minimal initial samples.

Contribution

The authors introduce an iterative active-learning approach that enhances traffic sign recognition, effectively utilizing low-confidence outputs to identify rare signs from large unlabeled datasets.

Findings

Efficient identification of rare traffic signs using active search.

Comparable results starting from a synthetic sample.

Improved recognition performance for automated driving systems.

Abstract

We describe an iterative active-learning algorithm to recognise rare traffic signs. A standard ResNet is trained on a training set containing only a single sample of the rare class. We demonstrate that by sorting the samples of a large, unlabeled set by the estimated probability of belonging to the rare class, we can efficiently identify samples from the rare class. This works despite the fact that this estimated probability is usually quite low. A reliable active-learning loop is obtained by labeling these candidate samples, including them in the training set, and iterating the procedure. Further, we show that we get similar results starting from a single synthetic sample. Our results are important as they indicate a straightforward way of improving traffic-sign recognition for automated driving systems. In addition, they show that we can make use of the information hidden in low confidence outputs, which is usually ignored.