Six-channel Image Representation for Cross-domain Object Detection

TL;DR

This paper introduces a novel six-channel image representation combining original and GAN-generated images to improve cross-domain object detection, addressing domain shift issues.

Contribution

The study proposes a new 6-channel image representation that leverages GAN-generated images to enhance cross-domain object detection performance.

Findings

Improved detection accuracy across different domains.

Effective utilization of GAN-generated data for domain adaptation.

Potential for inspiring further research on data augmentation techniques.

Abstract

Most deep learning models are data-driven and the excellent performance is highly dependent on the abundant and diverse datasets. However, it is very hard to obtain and label the datasets of some specific scenes or applications. If we train the detector using the data from one domain, it cannot perform well on the data from another domain due to domain shift, which is one of the big challenges of most object detection models. To address this issue, some image-to-image translation techniques have been employed to generate some fake data of some specific scenes to train the models. With the advent of Generative Adversarial Networks (GANs), we could realize unsupervised image-to-image translation in both directions from a source to a target domain and from the target to the source domain. In this study, we report a new approach to making use of the generated images. We propose to concatenate the original 3-channel images and their corresponding GAN-generated fake images to form 6-channel representations of the dataset, hoping to address the domain shift problem while exploiting the success of available detection models. The idea of augmented data representation may inspire further study on object detection and other applications.