{"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/neural-ctrl-f-segmentation-free-query-by","title":"Neural Ctrl-F: Segmentation-free Query-by-String Word Spotting in Handwritten Manuscript Collections","arxiv_id":"1703.07645","date":"2017-03-22","proceeding":"ICCV 2017 10","authors":["Tomas Wilkinson","Jonas Lindström","Anders Brun"],"abstract":"In this paper, we approach the problem of segmentation-free query-by-string\nword spotting for handwritten documents. In other words, we use methods\ninspired from computer vision and machine learning to search for words in large\ncollections of digitized manuscripts. In particular, we are interested in\nhistorical handwritten texts, which are often far more challenging than modern\nprinted documents. This task is important, as it provides people with a way to\nquickly find what they are looking for in large collections that are tedious\nand difficult to read manually. To this end, we introduce an end-to-end\ntrainable model based on deep neural networks that we call Ctrl-F-Net. Given a\nfull manuscript page, the model simultaneously generates region proposals, and\nembeds these into a distributed word embedding space, where searches are\nperformed. We evaluate the model on common benchmarks for handwritten word\nspotting, outperforming the previous state-of-the-art segmentation-free\napproaches by a large margin, and in some cases even segmentation-based\napproaches. One interesting real-life application of our approach is to help\nhistorians to find and count specific words in court records that are related\nto women's sustenance activities and division of labor. We provide promising\npreliminary experiments that validate our method on this task.","url_abs":"http://arxiv.org/abs/1703.07645v2","url_pdf":"http://arxiv.org/pdf/1703.07645v2.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":"neural-ctrl-f-segmentation-free-query-by","repo_url":"https://github.com/tomfalainen/neural-ctrlf","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"torch","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}