Cut-and-Paste Dataset Generation for Balancing Domain Gaps in Object Instance Detection

TL;DR

This paper proposes an advanced cut-and-paste dataset generation method that balances foreground and background domain gaps using GANs and image processing, improving object detection accuracy with limited data.

Contribution

It introduces a novel approach to balance unaligned domain gaps in cut-and-paste datasets by diversifying foregrounds with GANs and simplifying backgrounds, enhancing detection performance.

Findings

Improved object detection accuracy in cluttered environments.

Balancing domain gaps enhances the effectiveness of domain adaptation methods.

Effective with limited seed images.

Abstract

Training an object instance detector where only a few training object images are available is a challenging task. One solution is a cut-and-paste method that generates a training dataset by cutting object areas out of training images and pasting them onto other background images. A detector trained on a dataset generated with a cut-and-paste method suffers from the conventional domain shift problem, which stems from a discrepancy between the source domain (generated training dataset) and the target domain (real test dataset). Though state-of-the-art domain adaptation methods are able to reduce this gap, it is limited because they do not consider the difference of domain gaps of foreground and background. In this study, we present that the conventional domain gap can be divided into two sub-domain gaps for foreground and background. Then, we show that the original cut-and-paste approach suffers from a new domain gap problem, an unbalanced domain gaps, because it has two separate source domains for foreground and background, unlike the conventional domain shift problem. Then, we introduce an advanced cut-and-paste method to balance the unbalanced domain gaps by diversifying the foreground with GAN (generative adversarial network)-generated seed images and simplifying the background using image processing techniques. Experimental results show that our method is effective for balancing domain gaps and improving the accuracy of object instance detection in a cluttered indoor environment using only a few seed images. Furthermore, we show that balancing domain gaps can improve the detection accuracy of state-of-the-art domain adaptation methods.