{"id":13543025,"url":"https://github.com/vsymbol/CUTIE","last_synced_at":"2025-04-02T12:31:04.777Z","repository":{"id":37617696,"uuid":"165796281","full_name":"vsymbol/CUTIE","owner":"vsymbol","description":"CUTIE (TensorFlow implementation of Convolutional Universal Text Information Extractor)","archived":false,"fork":false,"pushed_at":"2022-12-08T05:25:04.000Z","size":3012,"stargazers_count":154,"open_issues_count":18,"forks_count":78,"subscribers_count":16,"default_branch":"master","last_synced_at":"2024-11-03T09:33:38.102Z","etag":null,"topics":["computer-vision","deep-learning","text-extraction"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/vsymbol.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-01-15T06:18:27.000Z","updated_at":"2024-10-05T15:00:23.000Z","dependencies_parsed_at":"2023-01-24T12:30:27.872Z","dependency_job_id":null,"html_url":"https://github.com/vsymbol/CUTIE","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vsymbol%2FCUTIE","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vsymbol%2FCUTIE/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vsymbol%2FCUTIE/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/vsymbol%2FCUTIE/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/vsymbol","download_url":"https://codeload.github.com/vsymbol/CUTIE/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246815392,"owners_count":20838435,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["computer-vision","deep-learning","text-extraction"],"created_at":"2024-08-01T11:00:21.637Z","updated_at":"2025-04-02T12:30:59.768Z","avatar_url":"https://github.com/vsymbol.png","language":"Python","funding_links":[],"categories":["Text detection and localization"],"sub_categories":["Form Segmentation"],"readme":"# CUTIE\nTensorFlow implementation of the paper \"CUTIE: Learning to Understand Documents with Convolutional Universal Text Information Extractor.\"\nXiaohui Zhao [Paper Link](https://arxiv.org/abs/1903.12363v4)\n\n----\nCUTIE 是用于“票据文档” 2D 关键信息提取/命名实体识别/槽位填充 算法。\n使用CUTIE前，需先使用OCR算法对“票据文档” 中的文字执行检测和识别，而后将格式化的文本输入入CUTIE网络，具体流程可参照论文。\n\nCUTIE can be considered as one type of 2-Dimensional Key Information Extraction, 2-D NER (Named Entity Recognition) or a 2-Dimensional 2D Slot Filling algorithm.\nBefore training / inference with CUTIE, prepare your structured texts in your scanned document images with any type of OCR algorithm. Refer to the CUTIE paper for details about the procedure.\n\n### Results\n\nResult evaluated on 4,484 receipt documents, including taxi receipts, meals entertainment receipts, and hotel receipts, with 9 different key information classes. (AP / softAP)\n|Method     | #Params   |  Taxi         |  Hotel        |\n| ----------|:---------:| :-----:       | :-----:       |\n| CloudScan | -         |  82.0 / -     |  60.0 / -     |\n| BERT      | 110M      |  88.1 / -     |  71.7 / -     |\n| CUTIE     |**14M**    |**94.0 / 97.3**|**74.6 / 87.0**|\n\n![Taxi](https://github.com/vsymbol/CUTIE/raw/master/others/example_1.jpg)\n\n![Hotel](https://github.com/vsymbol/CUTIE/raw/master/others/example_2.jpg)\n\n\n### Installation \u0026 Usage\n\n```\npip install -r requirements.txt\n```\n\n1. Generate your own dictionary with main_build_dict.py / main_data_tokenizer.py\n2. Train your model with main_train_json.py\n\nCUTIE achieves best performance with rows/cols well configured. For more insights, refer to statistics in the file (others/TrainingStatistic.xlsx).\n\n![Chart](https://github.com/vsymbol/CUTIE/raw/master/others/chart.jpg)\n\n\n### Others\n\nFor information about the input example, refer to [issue discussion](https://github.com/vsymbol/CUTIE/issues/7).\n- Apply any OCR tool that help you detecting and recognizing words in the scanned document image.\n- Label image OCR results with key information class as the .json file in the invoice_data folder. (thanks to @4kssoft)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvsymbol%2FCUTIE","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvsymbol%2FCUTIE","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvsymbol%2FCUTIE/lists"}