This article is cited in 11 papers

IMAGE PROCESSING, PATTERN RECOGNITION

Road images augmentation with synthetic traffic signs using neural networks

A. S. Konushin, B. V. Faizov, V. I. Shakhuro

Lomonosov Moscow State University, Faculty of Mechanics and Mathematics

Abstract: Traffic sign recognition is a well-researched problem in computer vision. However, the state of the art methods works only for frequent sign classes, which are well represented in training datasets. We consider the task of rare traffic sign detection and classification. We aim to solve that problem by using synthetic training data. Such training data is obtained by embedding synthetic images of signs in the real photos. We propose three methods for making synthetic signs consistent with a scene in appearance. These methods are based on modern generative adversarial network (GAN) architectures. Our proposed methods allow realistic embedding of rare traffic sign classes that are absent in the training set. We adapt a variational autoencoder for sampling plausible locations of new traffic signs in images. We demonstrate that using a mixture of our synthetic data with real data improves the accuracy of both classifier and detector.

Keywords: traffic sign classification, synthetic training samples, neural networks, image recognition, image transforms, neural network compositions

Received: 30.12.2020
Accepted: 19.04.2021

DOI: 10.18287/2412-6179-CO-859