{"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/end-to-end-weakly-supervised-semantic","title":"End-to-end weakly-supervised semantic alignment","arxiv_id":"1712.06861","date":"2017-12-19","proceeding":"CVPR 2018 6","authors":["Ignacio Rocco","Relja Arandjelović","Josef Sivic"],"abstract":"We tackle the task of semantic alignment where the goal is to compute dense\nsemantic correspondence aligning two images depicting objects of the same\ncategory. This is a challenging task due to large intra-class variation,\nchanges in viewpoint and background clutter. We present the following three\nprincipal contributions. First, we develop a convolutional neural network\narchitecture for semantic alignment that is trainable in an end-to-end manner\nfrom weak image-level supervision in the form of matching image pairs. The\noutcome is that parameters are learnt from rich appearance variation present in\ndifferent but semantically related images without the need for tedious manual\nannotation of correspondences at training time. Second, the main component of\nthis architecture is a differentiable soft inlier scoring module, inspired by\nthe RANSAC inlier scoring procedure, that computes the quality of the alignment\nbased on only geometrically consistent correspondences thereby reducing the\neffect of background clutter. Third, we demonstrate that the proposed approach\nachieves state-of-the-art performance on multiple standard benchmarks for\nsemantic alignment.","url_abs":"http://arxiv.org/abs/1712.06861v2","url_pdf":"http://arxiv.org/pdf/1712.06861v2.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":"end-to-end-weakly-supervised-semantic","repo_url":"https://github.com/hukim1112/weakalign","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"end-to-end-weakly-supervised-semantic","repo_url":"https://github.com/ignacio-rocco/weakalign","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"semantic-correspondence","task_name":"Semantic correspondence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.06861","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}