A full data augmentation pipeline for small object detection based on generative adversarial networks

Object detection accuracy on small objects, i.e., objects under 32 x 32 pixels, lags behind that of large ones. To address this issue, innovative architectures have been designed and new datasets have been re-leased. Still, the number of small objects in many datasets does not suffice for training. The advent of the generative adversarial networks (GANs) opens up a new data augmentation possibility for training architectures without the costly task of annotating huge datasets for small objects. In this paper, we pro-pose a full pipeline for data augmentation for small object detection which combines a GAN-based object generator with techniques of object segmentation, image inpainting, and image blending to achieve high -quality synthetic data. The main component of our pipeline is DS-GAN, a novel GAN-based architecture that generates realistic small objects from larger ones. Experimental results show that our overall data augmentation method improves the performance of state-of-the-art models up to 11.9% AP@.5 s on UAVDT and by 4.7% AP@.5 s on iSAID, both for the small objects subset and for a scenario where the number of training instances is limited.(c) 2022 The Author(s). Published by Elsevier Ltd.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)