{"url":"/task/handwritten-text-recognition","name":"Handwritten Text Recognition","slug":"handwritten-text-recognition","description_markdown":"Handwritten Text Recognition (HTR) is the task of automatically identifying and transcribing handwritten text from images or scanned documents into machine-readable text. The goal is to develop a system capable of accurately interpreting diverse handwriting styles, accounting for variations in alignment, stroke, spacing, and noise. This task involves detecting handwritten regions within an image, extracting the text content, and converting it into a structured digital format, enabling further search, indexing, or data analysis.","categories":[{"name":"Adversarial","url":"/area/adversarial"},{"name":"Computer Vision","url":"/area/computer-vision"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":139,"papers_with_code":56,"benchmarks":13,"benchmark_tables_in_archive":13,"benchmark_tables_shown":13,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":14,"subtasks":2,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/handwritten-text-recognition-on-iam","slug":"handwritten-text-recognition-on-iam","dataset":"IAM","dataset_url":"/dataset/iam","rows_in_archive":17,"metrics":["CER","WER"],"first_row_in_archive_order":{"model":"DTrOCR 105M","paper_title":"DTrOCR: Decoder-only Transformer for Optical Character Recognition","paper_url":"/paper/dtrocr-decoder-only-transformer-for-optical","paper_date":"2023-08-30","arxiv_id":"2308.15996","code_links":[{"title":"arvindrajan92/DTrOCR","url":"https://github.com/arvindrajan92/DTrOCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-lam-line","slug":"handwritten-text-recognition-on-lam-line","dataset":"LAM(line-level)","dataset_url":"/dataset/lam-line-level","rows_in_archive":6,"metrics":["Test CER","Test WER"],"first_row_in_archive_order":{"model":"HTR-VT","paper_title":"HTR-VT: Handwritten Text Recognition with Vision Transformer","paper_url":"/paper/htr-vt-handwritten-text-recognition-with","paper_date":"2024-09-13","arxiv_id":"2409.08573","code_links":[{"title":"yutingli0606/htr-vt","url":"https://github.com/yutingli0606/htr-vt"},{"title":"mortezagolzan/Hand_Writing_Text_Extarction","url":"https://github.com/mortezagolzan/Hand_Writing_Text_Extarction"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-iam-line","slug":"handwritten-text-recognition-on-iam-line","dataset":"IAM(line-level)","dataset_url":"/dataset/iam-line-level","rows_in_archive":5,"metrics":["Test CER","Test WER"],"first_row_in_archive_order":{"model":"TrOCR","paper_title":"TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models","paper_url":"/paper/trocr-transformer-based-optical-character","paper_date":"2021-09-21","arxiv_id":"2109.10282","code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"microsoft/unilm","url":"https://github.com/microsoft/unilm/tree/master/trocr"},{"title":"oleehyo/texteller","url":"https://github.com/oleehyo/texteller"},{"title":"d-gurgurov/im2latex","url":"https://github.com/d-gurgurov/im2latex"},{"title":"prathameshza/TrOCR_FineTuning","url":"https://github.com/prathameshza/TrOCR_FineTuning"},{"title":"MindCode-4/code-5","url":"https://github.com/MindCode-4/code-5/tree/main/trocr"},{"title":"pwc-1/Paper-9","url":"https://github.com/pwc-1/Paper-9/tree/main/1/trocr"},{"title":"pwc-1/Paper-10","url":"https://github.com/pwc-1/Paper-10/tree/main/trocr"}],"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}}},{"leaderboard":"/sota/handwritten-text-recognition-on-read2016-line","slug":"handwritten-text-recognition-on-read2016-line","dataset":"READ2016(line-level)","dataset_url":"/dataset/read2016-line-level","rows_in_archive":5,"metrics":["Test CER","Test WER"],"first_row_in_archive_order":{"model":"HTR-VT","paper_title":"HTR-VT: Handwritten Text Recognition with Vision Transformer","paper_url":"/paper/htr-vt-handwritten-text-recognition-with","paper_date":"2024-09-13","arxiv_id":"2409.08573","code_links":[{"title":"yutingli0606/htr-vt","url":"https://github.com/yutingli0606/htr-vt"},{"title":"mortezagolzan/Hand_Writing_Text_Extarction","url":"https://github.com/mortezagolzan/Hand_Writing_Text_Extarction"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-belfort","slug":"handwritten-text-recognition-on-belfort","dataset":"Belfort","dataset_url":"/dataset/belfort","rows_in_archive":4,"metrics":["CER (%)","WER (%)"],"first_row_in_archive_order":{"model":"PyLaia (all transcriptions + agreement-based split)","paper_title":"Handwritten Text Recognition from Crowdsourced Annotations","paper_url":"/paper/handwritten-text-recognition-from-1","paper_date":"2023-06-19","arxiv_id":"2306.10878","code_links":[],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-read-2016","slug":"handwritten-text-recognition-on-read-2016","dataset":"READ 2016","dataset_url":"/dataset/read-2016","rows_in_archive":2,"metrics":["CER (%)","WER (%)"],"first_row_in_archive_order":{"model":"DAN","paper_title":"DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognition","paper_url":"/paper/dan-a-segmentation-free-document-attention","paper_date":"2022-03-23","arxiv_id":"2203.12273","code_links":[{"title":"factodeeplearning/dan","url":"https://github.com/factodeeplearning/dan"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-bentham","slug":"handwritten-text-recognition-on-bentham","dataset":"Bentham","dataset_url":"/dataset/bentham","rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-digital-peter","slug":"handwritten-text-recognition-on-digital-peter","dataset":"Digital Peter","dataset_url":"/dataset/digital-peter","rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-hkr","slug":"handwritten-text-recognition-on-hkr","dataset":"HKR","dataset_url":"/dataset/hkr","rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-iam-b","slug":"handwritten-text-recognition-on-iam-b","dataset":"IAM-B","dataset_url":null,"rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-iam-d","slug":"handwritten-text-recognition-on-iam-d","dataset":"IAM-D","dataset_url":null,"rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-saint-gall","slug":"handwritten-text-recognition-on-saint-gall","dataset":"Saint Gall","dataset_url":"/dataset/saint-gall","rows_in_archive":1,"metrics":["CER"],"first_row_in_archive_order":{"model":"StackMix+Blots","paper_title":"StackMix and Blot Augmentations for Handwritten Text Recognition","paper_url":"/paper/stackmix-and-blot-augmentations-for","paper_date":"2021-08-26","arxiv_id":"2108.11667","code_links":[{"title":"sberbank-ai/StackMix-OCR","url":"https://github.com/sberbank-ai/StackMix-OCR"}],"syntology":null}},{"leaderboard":"/sota/handwritten-text-recognition-on-simara","slug":"handwritten-text-recognition-on-simara","dataset":"SIMARA","dataset_url":"/dataset/simara","rows_in_archive":1,"metrics":["CER (%)","WER (%)"],"first_row_in_archive_order":{"model":"DAN","paper_title":"SIMARA: a database for key-value information extraction from full pages","paper_url":"/paper/simara-a-database-for-key-value-information","paper_date":"2023-04-26","arxiv_id":"2304.13606","code_links":[],"syntology":null}}],"datasets":[{"url":"/dataset/iam","name":"IAM","full_name":"IAM Handwriting","num_papers_in_archive":198},{"url":"/dataset/read2016-line-level","name":"READ2016(line-level)","full_name":"Line-level Handwritten Text Recognition on READ 2016","num_papers_in_archive":6},{"url":"/dataset/iam-line-level","name":"IAM(line-level)","full_name":"Line-level Handwritten Text Recognition on IAM","num_papers_in_archive":5},{"url":"/dataset/read-2016","name":"READ 2016","full_name":"HTR Dataset ICFHR 2016","num_papers_in_archive":5},{"url":"/dataset/bentham","name":"Bentham","full_name":"Bentham project","num_papers_in_archive":4},{"url":"/dataset/hkr","name":"HKR","full_name":"Handwritten Kazakh and Russian (HKR) Database for Text Recognition","num_papers_in_archive":4},{"url":"/dataset/lam-line-level","name":"LAM(line-level)","full_name":"The LAM Dataset: A Novel Benchmark for Line-Level Handwritten Text Recognition","num_papers_in_archive":4},{"url":"/dataset/digital-peter","name":"Digital Peter","full_name":"","num_papers_in_archive":3},{"url":"/dataset/bn-htrd","name":"BN-HTRd","full_name":"BN-HTRd: A Benchmark Dataset for Document Level Offline Bangla Handwritten Text Recognition (HTR)","num_papers_in_archive":2},{"url":"/dataset/saint-gall","name":"Saint Gall","full_name":"","num_papers_in_archive":2},{"url":"/dataset/simara","name":"SIMARA","full_name":"SIMARA: a database for key-value information extraction from full-page handwritten documents","num_papers_in_archive":2},{"url":"/dataset/belfort","name":"Belfort","full_name":"The Belfort dataset: Handwritten Text Recognition from Crowdsourced Annotations","num_papers_in_archive":1},{"url":"/dataset/matrivasha","name":"MatriVasha:","full_name":"MatriVasha: Compound Character atasetD","num_papers_in_archive":1},{"url":"/dataset/burmese-handwritten-digit-dataset-bhdd","name":"Burmese Handwritten Digit Dataset (BHDD)","full_name":"","num_papers_in_archive":0}],"subtasks":[{"url":"/task/handwritten-document-recognition","name":"Handwritten Document Recognition"},{"url":"/task/unsupervised-text-recognition","name":"Unsupervised Text Recognition"}],"parent_tasks":[{"url":"/task/optical-character-recognition","name":"Optical Character Recognition (OCR)"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":56,"tagged_in_all":139,"items":[{"url":"/paper/trocr-transformer-based-optical-character","title":"TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models","date":"2021-09-21","arxiv_id":"2109.10282","repositories_listed":8,"syntology":{"n":6,"n_ran":0,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/origaminet-weakly-supervised-segmentation-1","title":"OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold","date":"2020-06-12","arxiv_id":"2006.07491","repositories_listed":8,"syntology":null},{"url":"/paper/full-page-handwriting-recognition-via-image","title":"Full Page Handwriting Recognition via Image to Sequence Extraction","date":"2021-03-11","arxiv_id":"2103.06450","repositories_listed":5,"syntology":null},{"url":"/paper/decoupled-attention-network-for-text","title":"Decoupled Attention Network for Text Recognition","date":"2019-12-21","arxiv_id":"1912.10205","repositories_listed":5,"syntology":null},{"url":"/paper/scrabblegan-semi-supervised-varying-length","title":"ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation","date":"2020-03-23","arxiv_id":"2003.10557","repositories_listed":3,"syntology":{"n":15,"n_ran":2,"n_unverified":13,"n_pointer_only":0}},{"url":"/paper/htr-vt-handwritten-text-recognition-with","title":"HTR-VT: Handwritten Text Recognition with Vision Transformer","date":"2024-09-13","arxiv_id":"2409.08573","repositories_listed":2,"syntology":null},{"url":"/paper/muharaf-manuscripts-of-handwritten-arabic","title":"Muharaf: Manuscripts of Handwritten Arabic Dataset for Cursive Text Recognition","date":"2024-06-13","arxiv_id":"2406.09630","repositories_listed":2,"syntology":null},{"url":"/paper/data-generation-for-post-ocr-correction-of","title":"Data Generation for Post-OCR correction of Cyrillic handwriting","date":"2023-11-27","arxiv_id":"2311.15896","repositories_listed":2,"syntology":null},{"url":"/paper/marginalia-and-machine-learning-handwritten","title":"Uncovering the Handwritten Text in the Margins: End-to-end Handwritten Text Detection and Recognition","date":"2023-03-10","arxiv_id":"2303.05929","repositories_listed":2,"syntology":null},{"url":"/paper/digital-peter-dataset-competition-and","title":"Digital Peter: Dataset, Competition and Handwriting Recognition Methods","date":"2021-03-16","arxiv_id":"2103.09354","repositories_listed":2,"syntology":null},{"url":"/paper/sequence-to-sequence-contrastive-learning-for","title":"Sequence-to-Sequence Contrastive Learning for Text Recognition","date":"2020-12-20","arxiv_id":"2012.10873","repositories_listed":2,"syntology":null},{"url":"/paper/meta-dan-towards-an-efficient-prediction","title":"Meta-DAN: towards an efficient prediction strategy for page-level handwritten text recognition","date":"2025-04-04","arxiv_id":"2504.03349","repositories_listed":1,"syntology":null},{"url":"/paper/judge-a-book-by-its-cover-investigating-multi","title":"Judge a Book by its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document Transcription","date":"2025-02-27","arxiv_id":"2502.20295","repositories_listed":1,"syntology":null},{"url":"/paper/unlocking-the-archives-using-large-language","title":"Unlocking the Archives: Using Large Language Models to Transcribe Handwritten Historical Documents","date":"2024-11-02","arxiv_id":"2411.03340","repositories_listed":1,"syntology":null},{"url":"/paper/hespi-a-pipeline-for-automatically-detecting","title":"Hespi: A pipeline for automatically detecting information from hebarium specimen sheets","date":"2024-10-11","arxiv_id":"2410.08740","repositories_listed":1,"syntology":null},{"url":"/paper/platypus-a-generalized-specialist-model-for","title":"Platypus: A Generalized Specialist Model for Reading Text in Various Forms","date":"2024-08-27","arxiv_id":"2408.14805","repositories_listed":1,"syntology":null},{"url":"/paper/daniel-a-fast-document-attention-network-for","title":"DANIEL: A fast Document Attention Network for Information Extraction and Labelling of handwritten documents","date":"2024-07-12","arxiv_id":"2407.09103","repositories_listed":1,"syntology":null},{"url":"/paper/best-practices-for-a-handwritten-text","title":"Best Practices for a Handwritten Text Recognition System","date":"2024-04-17","arxiv_id":"2404.11339","repositories_listed":1,"syntology":null},{"url":"/paper/automatic-transcription-of-handwritten-old","title":"Automatic Transcription of Handwritten Old Occitan Language","date":"2023-12-06","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/exploring-ocr-capabilities-of-gpt-4v-ision-a","title":"Exploring OCR Capabilities of GPT-4V(ision) : A Quantitative and In-depth Evaluation","date":"2023-10-25","arxiv_id":"2310.16809","repositories_listed":1,"syntology":{"n":4,"n_ran":4,"n_unverified":0,"n_pointer_only":4}},{"url":"/paper/dtrocr-decoder-only-transformer-for-optical","title":"DTrOCR: Decoder-only Transformer for Optical Character Recognition","date":"2023-08-30","arxiv_id":"2308.15996","repositories_listed":1,"syntology":null},{"url":"/paper/handwritten-stenography-recognition-and-the","title":"Handwritten Stenography Recognition and the LION Dataset","date":"2023-08-15","arxiv_id":"2308.07799","repositories_listed":1,"syntology":null},{"url":"/paper/writer-adaptation-for-offline-text","title":"Writer adaptation for offline text recognition: An exploration of neural network-based methods","date":"2023-07-11","arxiv_id":"2307.15071","repositories_listed":1,"syntology":null},{"url":"/paper/bn-drishti-bangla-document-recognition","title":"BN-DRISHTI: Bangla Document Recognition through Instance-level Segmentation of Handwritten Text Images","date":"2023-05-31","arxiv_id":"2306.09351","repositories_listed":1,"syntology":null},{"url":"/paper/faster-dan-multi-target-queries-with-document","title":"Faster DAN: Multi-target Queries with Document Positional Encoding for End-to-end Handwritten Document Recognition","date":"2023-01-25","arxiv_id":"2301.10593","repositories_listed":1,"syntology":null},{"url":"/paper/end-to-end-page-level-assessment-of","title":"End-to-End Page-Level Assessment of Handwritten Text Recognition","date":"2023-01-14","arxiv_id":"2301.05935","repositories_listed":1,"syntology":null},{"url":"/paper/enhancing-indic-handwritten-text-recognition","title":"Enhancing Indic Handwritten Text Recognition Using Global Semantic Information","date":"2022-12-15","arxiv_id":"2212.07776","repositories_listed":1,"syntology":null},{"url":"/paper/towards-end-to-end-handwritten-document","title":"Towards End-to-end Handwritten Document Recognition","date":"2022-09-30","arxiv_id":"2209.15362","repositories_listed":1,"syntology":null},{"url":"/paper/handwritten-paleographic-greek-text","title":"Handwritten Paleographic Greek Text Recognition: A Century-Based Approach","date":"2022-06-01","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/easter2-0-improving-convolutional-models-for","title":"Easter2.0: Improving convolutional models for handwritten text recognition","date":"2022-05-30","arxiv_id":"2205.14879","repositories_listed":1,"syntology":null}],"syntology_records":3,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}