{"url":"/sota/scene-text-recognition-on-svtp","task":{"name":"Scene Text Recognition","url":"/task/scene-text-recognition","note":null},"dataset":{"name":"SVTP","url":"/dataset/svtp"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"See [Scene Text Detection](https://paperswithcode.com/task/scene-text-detection) for leaderboards in this task.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy":"higher"}},"counts":{"rows":17,"rows_with_code":16,"rows_with_paper_page":17,"rows_dated":17,"rows_using_additional_data":9},"rows":[{"rank_in_archive_order":1,"model":"DTrOCR 105M","metrics":{"Accuracy":"98.6"},"uses_additional_data":false,"paper_date":"2023-08-30","paper":"/paper/dtrocr-decoder-only-transformer-for-optical","paper_url":"https://arxiv.org/abs/2308.15996v1","paper_title":"DTrOCR: Decoder-only Transformer for Optical Character Recognition","code":"https://github.com/arvindrajan92/DTrOCR","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"MGP-STR","metrics":{"Accuracy":"98.3"},"uses_additional_data":true,"paper_date":"2022-09-08","paper":"/paper/multi-granularity-prediction-for-scene-text","paper_url":"https://arxiv.org/abs/2209.03592v2","paper_title":"Multi-Granularity Prediction for Scene Text Recognition","code":"https://github.com/alibabaresearch/advancedliteratemachinery","n_code_links":3,"syntology":null},{"rank_in_archive_order":3,"model":"CLIP4STR-L*","metrics":{"Accuracy":"98.13"},"uses_additional_data":true,"paper_date":"2023-12-29","paper":"/paper/an-empirical-study-of-scaling-law-for-ocr","paper_url":"https://arxiv.org/abs/2401.00028v3","paper_title":"An Empirical Study of Scaling Law for OCR","code":"https://github.com/large-ocr-model/large-ocr-model.github.io","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"CLIP4STR-L (DataComp-1B)","metrics":{"Accuracy":"98.1"},"uses_additional_data":false,"paper_date":"2023-05-23","paper":"/paper/clip4str-a-simple-baseline-for-scene-text-1","paper_url":"https://arxiv.org/abs/2305.14014v4","paper_title":"CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model","code":"https://github.com/VamosC/CLIP4STR","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"CLIP4STR-L","metrics":{"Accuracy":"97.4"},"uses_additional_data":false,"paper_date":"2023-05-23","paper":"/paper/clip4str-a-simple-baseline-for-scene-text-1","paper_url":"https://arxiv.org/abs/2305.14014v4","paper_title":"CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model","code":"https://github.com/VamosC/CLIP4STR","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"CLIP4STR-B","metrics":{"Accuracy":"97.2"},"uses_additional_data":true,"paper_date":"2023-05-23","paper":"/paper/clip4str-a-simple-baseline-for-scene-text-1","paper_url":"https://arxiv.org/abs/2305.14014v4","paper_title":"CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model","code":"https://github.com/VamosC/CLIP4STR","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"CPPD","metrics":{"Accuracy":"96.7"},"uses_additional_data":true,"paper_date":"2023-07-23","paper":"/paper/context-perception-parallel-decoder-for-scene","paper_url":"https://arxiv.org/abs/2307.12270v2","paper_title":"Context Perception Parallel Decoder for Scene Text Recognition","code":"https://github.com/PaddlePaddle/PaddleOCR","n_code_links":2,"syntology":null},{"rank_in_archive_order":8,"model":"CCD-ViT-Base","metrics":{"Accuracy":"96.1"},"uses_additional_data":true,"paper_date":"2022-11-01","paper":"/paper/self-supervised-character-to-character","paper_url":"https://arxiv.org/abs/2211.00288v4","paper_title":"Self-supervised Character-to-Character Distillation for Text Recognition","code":"https://github.com/tongkunguan/ccd","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"PARSeq","metrics":{"Accuracy":"95.7±0.9"},"uses_additional_data":true,"paper_date":"2022-07-14","paper":"/paper/scene-text-recognition-with-permuted","paper_url":"https://arxiv.org/abs/2207.06966v1","paper_title":"Scene Text Recognition with Permuted Autoregressive Sequence Models","code":"https://github.com/topdu/openocr","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"CCD-ViT-Small","metrics":{"Accuracy":"92.7"},"uses_additional_data":true,"paper_date":"2022-11-01","paper":"/paper/self-supervised-character-to-character","paper_url":"https://arxiv.org/abs/2211.00288v4","paper_title":"Self-supervised Character-to-Character Distillation for Text Recognition","code":"https://github.com/tongkunguan/ccd","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"CCD-ViT-Tiny","metrics":{"Accuracy":"91.6"},"uses_additional_data":true,"paper_date":"2022-11-01","paper":"/paper/self-supervised-character-to-character","paper_url":"https://arxiv.org/abs/2211.00288v4","paper_title":"Self-supervised Character-to-Character Distillation for Text Recognition","code":"https://github.com/tongkunguan/ccd","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"S-GTR","metrics":{"Accuracy":"90.6"},"uses_additional_data":true,"paper_date":"2021-12-24","paper":"/paper/visual-semantics-allow-for-textual-reasoning-1","paper_url":"https://arxiv.org/abs/2112.12916v1","paper_title":"Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition","code":"https://github.com/adeline-cs/GTR","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"MATRN","metrics":{"Accuracy":"90.6"},"uses_additional_data":false,"paper_date":"2021-11-30","paper":"/paper/multi-modal-text-recognition-networks","paper_url":"https://arxiv.org/abs/2111.15263v3","paper_title":"Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features","code":"https://github.com/topdu/openocr","n_code_links":3,"syntology":null},{"rank_in_archive_order":14,"model":"SIGA_T","metrics":{"Accuracy":"90.5"},"uses_additional_data":false,"paper_date":"2022-03-07","paper":"/paper/a-glyph-driven-topology-enhancement-network","paper_url":"https://arxiv.org/abs/2203.03382v4","paper_title":"Self-supervised Implicit Glyph Attention for Text Recognition","code":"https://github.com/tongkunguan/siga","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"CDistNet (Ours)","metrics":{"Accuracy":"89.77"},"uses_additional_data":false,"paper_date":"2021-11-22","paper":"/paper/cdistnet-perceiving-multi-domain-character","paper_url":"https://arxiv.org/abs/2111.11011v5","paper_title":"CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition","code":"https://github.com/topdu/openocr","n_code_links":3,"syntology":null},{"rank_in_archive_order":16,"model":"DiffusionSTR","metrics":{"Accuracy":"89.2"},"uses_additional_data":false,"paper_date":"2023-06-29","paper":"/paper/diffusionstr-diffusion-model-for-scene-text","paper_url":"https://arxiv.org/abs/2306.16707v1","paper_title":"DiffusionSTR: Diffusion Model for Scene Text Recognition","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":17,"model":"DPAN","metrics":{"Accuracy":"89.0"},"uses_additional_data":false,"paper_date":"2021-08-01","paper":"/paper/look-back-again-dual-parallel-attention","paper_url":"https://dl.acm.org/doi/10.1145/3460426.3463674","paper_title":"Look Back Again: Dual Parallel Attention Network for Accurate and Robust Scene Text Recognition","code":"https://github.com/siddagra/DPAN-look-back-Again-Dual-Parallel-Attention-Network-for-Accurate-and-Robust-Scene-Text-Recognition","n_code_links":2,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,795 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6795,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2785},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":4,"rows_with_any_sample_ran":4,"distinct_papers_with_graph_line":2,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":8,"n_unverified":10,"n_samples":18,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":14,"n_unverified":24,"n_samples":38,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}