Investigating Binary Neural Networks for Traffic Sign Detection and Recognition

Traffic sign detection is crucial for enabling autonomous vehicles to navigate in real-world streets, which must be carried out with high accuracy and in real-time. CNNs have become one of the standard approaches for traffic sign detection research in recent years. The use of CNNs has allowed the development of traffic sign detectors that are capable of achieving prediction accuracies similar to those of human drivers. However, most CNNs do not run in real-time due to the high number of computational operations involved during the inference phase. This hinders the deployment of CNNs in autonomous vehicles despite their high prediction accuracy. In this paper, we explore BNNs to tackle this problem. BNNs binarize the full-precision weights and activations of a CNN, drastically reducing the complexity of the computational operations required for inference, while at the same time maintaining the architectural parameters, as well as spatial dimensions of the input image. This reduces the memory required to run the model and enables faster inference time. We carry out in-depth studies on applying BNNs for traffic sign detection using real-world datasets. We observe an improvement of 11.63 × for normalized compute complexity, while suffering only 3.93 pp in detection accuracy on GTSDB dataset.

I. Introduction

Traffic signs are vital in maintaining road safety and controlling the flow of traffic [1]. This makes the ability to detect traffic signs an integral part of any vision system for autonomous driving. Building a traffic sign detector is challenging as it needs to cope with complex real-world traffic scenes with diverse background objects. This renders detectors using traditional computer vision algorithms, which rely on color and geometry of the traffic signs, unreliable and not robust enough, especially for urban scenarios. However, these challenges are gradually being tackled with the use of convolutional neural networks (CNNs).