Cross-Domain Image Classification through Neural-Style Transfer Data Augmentation

TL;DR

This paper investigates how style transfer data augmentation affects convolutional neural network accuracy in automobile detection under harsh winter weather conditions, highlighting its potential to improve robustness in noisy environments.

Contribution

It demonstrates the effectiveness of style transfer-based data augmentation for improving CNN performance in adverse weather conditions, a novel application in this context.

Findings

Style transfer augmentation improves detection accuracy in winter weather.

Classifiers trained with mixed datasets perform better on adverse weather images.

Style transfer augmentation has specific strengths and weaknesses discussed in the paper.

Abstract

In particular, the lack of sufficient amounts of domain-specific data can reduce the accuracy of a classifier. In this paper, we explore the effects of style transfer-based data transformation on the accuracy of a convolutional neural network classifiers in the context of automobile detection under adverse winter weather conditions. The detection of automobiles under highly adverse weather conditions is a difficult task as such conditions present large amounts of noise in each image. The InceptionV2 architecture is trained on a composite dataset, consisting of either normal car image dataset , a mixture of normal and style transferred car images, or a mixture of normal car images and those taken at blizzard conditions, at a ratio of 80:20. All three classifiers are then tested on a dataset of car images taken at blizzard conditions and on vehicle-free snow landscape images. We evaluate and contrast the effectiveness of each classifier upon each dataset, and discuss the strengths and weaknesses of style-transfer based approaches to data augmentation.