{"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/convolutional-crfs-for-semantic-segmentation","title":"Convolutional CRFs for Semantic Segmentation","arxiv_id":"1805.04777","date":"2018-05-12","proceeding":"ICLR 2019 5","authors":["Marvin T. T. Teichmann","Roberto Cipolla"],"abstract":"For the challenging semantic image segmentation task the most efficient\nmodels have traditionally combined the structured modelling capabilities of\nConditional Random Fields (CRFs) with the feature extraction power of CNNs. In\nmore recent works however, CRF post-processing has fallen out of favour. We\nargue that this is mainly due to the slow training and inference speeds of\nCRFs, as well as the difficulty of learning the internal CRF parameters. To\novercome both issues we propose to add the assumption of conditional\nindependence to the framework of fully-connected CRFs. This allows us to\nreformulate the inference in terms of convolutions, which can be implemented\nhighly efficiently on GPUs. Doing so speeds up inference and training by a\nfactor of more then 100. All parameters of the convolutional CRFs can easily be\noptimized using backpropagation. To facilitating further CRF research we make\nour implementation publicly available. Please visit:\nhttps://github.com/MarvinTeichmann/ConvCRF","url_abs":"http://arxiv.org/abs/1805.04777v2","url_pdf":"http://arxiv.org/pdf/1805.04777v2.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":"convolutional-crfs-for-semantic-segmentation","repo_url":"https://github.com/MarvinTeichmann/ConvCRF","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[{"method_slug":"crf","method_name":"CRF"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}