Can Synthetic Data Improve Object Detection Results for Remote Sensing Images?
arXiv:2006.05015
Abstract
Deep learning approaches require enough training samples to perform well, but it is a challenge to collect enough real training data and label them manually. In this letter, we propose the use of realistic synthetic data with a wide distribution to improve the performance of remote sensing image aircraft detection. Specifically, to increase the variability of synthetic data, we randomly set the parameters during rendering, such as the size of the instance and the class of background images. In order to make the synthetic images more realistic, we then refine the synthetic images at the pixel level using CycleGAN with real unlabeled images. We also fine-tune the model with a small amount of real data, to obtain a higher accuracy. Experiments on NWPU VHR-10, UCAS-AOD and DIOR datasets demonstrate that the proposed method can be applied for augmenting insufficient real data.
arXiv comment5 pages, 5 figures
References in corpus (4)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
- Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmark
- Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World