{"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/reading-text-in-the-wild-with-convolutional","title":"Reading Text in the Wild with Convolutional Neural Networks","arxiv_id":"1412.1842","date":"2014-12-04","proceeding":null,"authors":["Max Jaderberg","Karen Simonyan","Andrea Vedaldi","Andrew Zisserman"],"abstract":"In this work we present an end-to-end system for text spotting -- localising\nand recognising text in natural scene images -- and text based image retrieval.\nThis system is based on a region proposal mechanism for detection and deep\nconvolutional neural networks for recognition. Our pipeline uses a novel\ncombination of complementary proposal generation techniques to ensure high\nrecall, and a fast subsequent filtering stage for improving precision. For the\nrecognition and ranking of proposals, we train very large convolutional neural\nnetworks to perform word recognition on the whole proposal region at the same\ntime, departing from the character classifier based systems of the past. These\nnetworks are trained solely on data produced by a synthetic text generation\nengine, requiring no human labelled data.\n  Analysing the stages of our pipeline, we show state-of-the-art performance\nthroughout. We perform rigorous experiments across a number of standard\nend-to-end text spotting benchmarks and text-based image retrieval datasets,\nshowing a large improvement over all previous methods. Finally, we demonstrate\na real-world application of our text spotting system to allow thousands of\nhours of news footage to be instantly searchable via a text query.","url_abs":"http://arxiv.org/abs/1412.1842v1","url_pdf":"http://arxiv.org/pdf/1412.1842v1.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":[],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"region-proposal","task_name":"Region Proposal"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"text-spotting","task_name":"Text Spotting"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/scene-text-detection-on-icdar-2013","task":"Scene Text Detection","dataset":"ICDAR 2013","model":"Jaderberg et al.","rank_in_archive_order":15,"of":16,"metrics":{"F-Measure":"76.8%","Precision":"88.5","Recall":"67.8"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1412.1842","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}