{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/siftinggan-generating-and-sifting-labeled","title":"SiftingGAN: Generating and Sifting Labeled Samples to Improve the Remote Sensing Image Scene Classification Baseline in vitro","arxiv_id":"1809.04985","date":"2018-09-13","proceeding":null,"authors":["Dongao Ma","Ping Tang","Lijun Zhao"],"abstract":"Lack of annotated samples greatly restrains the direct application of deep\nlearning in remote sensing image scene classification. Although researches have\nbeen done to tackle this issue by data augmentation with various image\ntransformation operations, they are still limited in quantity and diversity.\nRecently, the advent of the unsupervised learning based generative adversarial\nnetworks (GANs) bring us a new way to generate augmented samples. However, such\nGAN-generated samples are currently only served for training GANs model itself\nand for improving the performance of the discriminator in GANs internally (in\nvivo). It becomes a question of serious doubt whether the GAN-generated samples\ncan help better improve the scene classification performance of other deep\nlearning networks (in vitro), compared with the widely used transformed\nsamples. To answer this question, this paper proposes a SiftingGAN approach to\ngenerate more numerous, more diverse and more authentic labeled samples for\ndata augmentation. SiftingGAN extends traditional GAN framework with an\nOnline-Output method for sample generation, a Generative-Model-Sifting method\nfor model sifting, and a Labeled-Sample-Discriminating method for sample\nsifting. Experiments on the well-known AID dataset demonstrate that the\nproposed SiftingGAN method can not only effectively improve the performance of\nthe scene classification baseline that is achieved without data augmentation,\nbut also significantly excels the comparison methods based on traditional\ngeometric/radiometric transformation operations.","url_abs":"http://arxiv.org/abs/1809.04985v4","url_pdf":"http://arxiv.org/pdf/1809.04985v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"siftinggan-generating-and-sifting-labeled","repo_url":"https://github.com/MaDongao/SiftingGAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"scene-classification","task_name":"Scene Classification"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}