{"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-scene-text-in-deep-convolutional","title":"Reading Scene Text in Deep Convolutional Sequences","arxiv_id":"1506.04395","date":"2015-06-14","proceeding":null,"authors":["Pan He","Weilin Huang","Yu Qiao","Chen Change Loy","Xiaoou Tang"],"abstract":"We develop a Deep-Text Recurrent Network (DTRN) that regards scene text\nreading as a sequence labelling problem. We leverage recent advances of deep\nconvolutional neural networks to generate an ordered high-level sequence from a\nwhole word image, avoiding the difficult character segmentation problem. Then a\ndeep recurrent model, building on long short-term memory (LSTM), is developed\nto robustly recognize the generated CNN sequences, departing from most existing\napproaches recognising each character independently. Our model has a number of\nappealing properties in comparison to existing scene text recognition methods:\n(i) It can recognise highly ambiguous words by leveraging meaningful context\ninformation, allowing it to work reliably without either pre- or\npost-processing; (ii) the deep CNN feature is robust to various image\ndistortions; (iii) it retains the explicit order information in word image,\nwhich is essential to discriminate word strings; (iv) the model does not depend\non pre-defined dictionary, and it can process unknown words and arbitrary\nstrings. Codes for the DTRN will be available.","url_abs":"http://arxiv.org/abs/1506.04395v2","url_pdf":"http://arxiv.org/pdf/1506.04395v2.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":"reading-scene-text-in-deep-convolutional","repo_url":"https://github.com/somitmittal/Reading-Scene-Text-from-Images-using-Tensorflow-CNN-Bidirectional-LSTM-CTC-Loss","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"scene-text-recognition","task_name":"Scene Text Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1506.04395","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}