RT Journal Article
T1 A full data augmentation pipeline for small object detection based on generative adversarial networks
A1 Bosquet Mera, Brais
A1 Cores Costa, Daniel
A1 Brea Sánchez, Víctor Manuel
A1 Mucientes Molina, Manuel
A1 Bimbo, Alberto del
K1 Small object detection
K1 Data augmentation
K1 Generative adversarial network
AB Object detection accuracy on small objects, i.e., objects under 32  32 pixels, lags behind that of large ones. To address this issue, innovative architectures have been designed and new datasets have been released. 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 propose 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 on UAVDT and by 4.7% AP on iSAID, both for the small objects subset and for a scenario where the number of training instances is limited.
PB Elsevier
YR 2023
FD 2023
LK http://hdl.handle.net/10347/29451
UL http://hdl.handle.net/10347/29451
LA eng
NO Pattern Recognition 133 2023 (108998)
NO This research was partially funded by the Spanish Ministerio de Ciencia e Innovación [grant numbers PID2020-112623GB-I00, RTI2018-097088-B-C32], and the Galician Consellería de Cultura, Educación e Universidade [grant numbers ED431C 2018/29, ED431C 2021/048, ED431G 2019/04]. These grants are co-funded by the European Regional Development Fund (ERDF). This paper was supported by European Union’s Horizon 2020 research and innovation programme under grant number
DS Minerva
RD 24 abr 2026