{"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/dense-haze-a-benchmark-for-image-dehazing","title":"Dense Haze: A benchmark for image dehazing with dense-haze and haze-free images","arxiv_id":"1904.02904","date":"2019-04-05","proceeding":null,"authors":["Codruta O. Ancuti","Cosmin Ancuti","Mateu Sbert","Radu Timofte"],"abstract":"Single image dehazing is an ill-posed problem that has recently drawn\nimportant attention. Despite the significant increase in interest shown for\ndehazing over the past few years, the validation of the dehazing methods\nremains largely unsatisfactory, due to the lack of pairs of real hazy and\ncorresponding haze-free reference images. To address this limitation, we\nintroduce Dense-Haze - a novel dehazing dataset. Characterized by dense and\nhomogeneous hazy scenes, Dense-Haze contains 33 pairs of real hazy and\ncorresponding haze-free images of various outdoor scenes. The hazy scenes have\nbeen recorded by introducing real haze, generated by professional haze\nmachines. The hazy and haze-free corresponding scenes contain the same visual\ncontent captured under the same illumination parameters. Dense-Haze dataset\naims to push significantly the state-of-the-art in single-image dehazing by\npromoting robust methods for real and various hazy scenes. We also provide a\ncomprehensive qualitative and quantitative evaluation of state-of-the-art\nsingle image dehazing techniques based on the Dense-Haze dataset. Not\nsurprisingly, our study reveals that the existing dehazing techniques perform\npoorly for dense homogeneous hazy scenes and that there is still much room for\nimprovement.","url_abs":"http://arxiv.org/abs/1904.02904v1","url_pdf":"http://arxiv.org/pdf/1904.02904v1.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":"dense-haze-a-benchmark-for-image-dehazing","repo_url":"https://github.com/pmm09c/ntire-dehazing","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"dense-haze-a-benchmark-for-image-dehazing","repo_url":"https://github.com/tianwenzhou/prodehaze","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"dense-haze-a-benchmark-for-image-dehazing","repo_url":"https://github.com/w-jilly/frequency-compensated-diffusion-model-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-dehazing","task_name":"Image Dehazing"},{"task_slug":"single-image-dehazing","task_name":"Single Image Dehazing"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.02904","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}