{"url":"/method/de-gan","slug":"de-gan","name":"DE-GAN","full_name":"DE-GAN: A Conditional Generative Adversarial Network for Document Enhancement","full_name_withheld":false,"description_markdown":"Documents often exhibit various forms of degradation, which make it hard to be read and substantially deteriorate the\r\nperformance of an OCR system. In this paper, we propose an effective end-to-end framework named Document Enhancement\r\nGenerative Adversarial Networks (DE-GAN) that uses the conditional GANs (cGANs) to restore severely degraded document images.\r\nTo the best of our knowledge, this practice has not been studied within the context of generative adversarial deep networks. We\r\ndemonstrate that, in different tasks (document clean up, binarization, deblurring and watermark removal), DE-GAN can produce an\r\nenhanced version of the degraded document with a high quality. In addition, our approach provides consistent improvements compared to state-of-the-art methods over the widely used DIBCO 2013, DIBCO 2017 and H-DIBCO 2018 datasets, proving its ability to restore a degraded document image to its ideal condition. The obtained results on a wide variety of degradation reveal the flexibility of the proposed model to be exploited in other document enhancement problems.","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":null,"title":null,"url_on_a_paper_host":false},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Generative Adversarial Networks","url":"/methods/category/generative-adversarial-networks","pwc_aliases":[]}],"n_papers_tagged":4,"archive_num_papers":4,"papers_newest_first":[{"paper":"/paper/a-fair-evaluation-of-various-deep-learning","title":"A Fair Evaluation of Various Deep Learning-Based Document Image Binarization Approaches","date":"2024-01-22","arxiv_id":"2401.11831","n_code_links":1,"syntology":null},{"paper":"/paper/de-gan-a-conditional-generative-adversarial-1","title":"DE-GAN: A Conditional Generative Adversarial Network for Document Enhancement","date":"2020-10-17","arxiv_id":"2010.08764","n_code_links":4,"syntology":null},{"paper":null,"title":"Facial Expression Representation Learning by Synthesizing Expression Images","date":"2019-11-30","arxiv_id":"1912.01456","n_code_links":0,"syntology":null},{"paper":null,"title":"Facial Expression Recognition Using Disentangled Adversarial Learning","date":"2019-09-28","arxiv_id":"1909.13135","n_code_links":0,"syntology":null}],"papers_shown":4,"tasks":[{"task":null,"name":"Generative Adversarial Network","papers":3},{"task":"/task/binarization","name":"Binarization","papers":2},{"task":"/task/decoder","name":"Decoder","papers":2},{"task":"/task/facial-expression-recognition-1","name":"Facial Expression Recognition","papers":2},{"task":"/task/facial-expression-recognition","name":"Facial Expression Recognition (FER)","papers":2},{"task":"/task/representation-learning","name":"Representation Learning","papers":2},{"task":"/task/deblurring","name":"Deblurring","papers":1},{"task":"/task/deep-learning","name":"Deep Learning","papers":1},{"task":"/task/document-enhancement","name":"Document Enhancement","papers":1},{"task":"/task/image-reconstruction","name":"Image Reconstruction","papers":1},{"task":"/task/optical-character-recognition","name":"Optical Character Recognition (OCR)","papers":1},{"task":null,"name":"valid","papers":1}],"tasks_shown":12,"n_tasks":12,"usage_by_year":[{"year":"2019","papers":2},{"year":"2020","papers":1},{"year":"2024","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/de-gan"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}