{"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/what-is-wrong-with-scene-text-recognition","title":"What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis","arxiv_id":"1904.01906","date":"2019-04-03","proceeding":"ICCV 2019 10","authors":["Jeonghun Baek","Geewook Kim","Junyeop Lee","Sungrae Park","Dongyoon Han","Sangdoo Yun","Seong Joon Oh","Hwalsuk Lee"],"abstract":"Many new proposals for scene text recognition (STR) models have been introduced in recent years. While each claim to have pushed the boundary of the technology, a holistic and fair comparison has been largely missing in the field due to the inconsistent choices of training and evaluation datasets. This paper addresses this difficulty with three major contributions. First, we examine the inconsistencies of training and evaluation datasets, and the performance gap results from inconsistencies. Second, we introduce a unified four-stage STR framework that most existing STR models fit into. Using this framework allows for the extensive evaluation of previously proposed STR modules and the discovery of previously unexplored module combinations. Third, we analyze the module-wise contributions to performance in terms of accuracy, speed, and memory demand, under one consistent set of training and evaluation datasets. Such analyses clean up the hindrance on the current comparisons to understand the performance gain of the existing modules.","url_abs":"https://arxiv.org/abs/1904.01906v4","url_pdf":"https://arxiv.org/pdf/1904.01906v4.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":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/clovaai/deep-text-recognition-benchmark","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/2023-MindSpore-1/ms-code-58","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/JoegameZhou/cnnctc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/Luckygyana/Invo-AI","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/RobotMobile/cv-deep-learning-paper-review","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/dmitrijsk/attentionhtr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/jakartaresearch/receipt-ocr","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/luozhilin99/cnnctc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/roatienza/deep-text-recognition-benchmark","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/2023-MindSpore-4/Code8/tree/main/cnnctc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/2024-MindSpore-1/Code7/tree/main/cnnctc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"what-is-wrong-with-scene-text-recognition","repo_url":"https://github.com/mindspore-ai/models/tree/master/official/cv/cnnctc","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"image-matching","task_name":"Image Matching"},{"task_slug":"scene-text-recognition","task_name":"Scene Text Recognition"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/scene-text-recognition-on-icdar-2003","task":"Scene Text Recognition","dataset":"ICDAR 2003","model":"Baek et al.","rank_in_archive_order":7,"of":12,"metrics":{"Accuracy":"94.4"},"uses_additional_data":false},{"leaderboard":"/sota/scene-text-recognition-on-icdar2013","task":"Scene Text Recognition","dataset":"ICDAR2013","model":"Baek et al.","rank_in_archive_order":31,"of":38,"metrics":{"Accuracy":"92.3"},"uses_additional_data":false},{"leaderboard":"/sota/scene-text-recognition-on-icdar2015","task":"Scene Text Recognition","dataset":"ICDAR2015","model":"Baek et al.","rank_in_archive_order":26,"of":27,"metrics":{"Accuracy":"71.8"},"uses_additional_data":false},{"leaderboard":"/sota/scene-text-recognition-on-svt","task":"Scene Text Recognition","dataset":"SVT","model":"Baek et al.","rank_in_archive_order":31,"of":37,"metrics":{"Accuracy":"87.5"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.01906","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1904.01906"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-4/Code8/tree/main/cnnctc","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Luckygyana/Invo-AI","reach":{"status":"ok","spdx":"GPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/clovaai/deep-text-recognition-benchmark","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2024-MindSpore-1/Code7/tree/main/cnnctc","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/luozhilin99/cnnctc","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/RobotMobile/cv-deep-learning-paper-review","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/2023-MindSpore-1/ms-code-58","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mindspore-ai/models/tree/master/official/cv/cnnctc","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/dmitrijsk/attentionhtr","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jakartaresearch/receipt-ocr","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/roatienza/deep-text-recognition-benchmark","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/JoegameZhou/cnnctc","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":19},"by_repo_kind":{"listed":{"samples":19,"ran":0,"repositories":4}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"316fb42e9e6ed79e","entry":"a_cosine_learning_rate","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/lr_schedule.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/lr_schedule.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"316fb42e9e6ed79e"}},{"code_sha256_prefix":"89c58f89ba416399","entry":"calcul_acc","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"postprocess.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/postprocess.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"89c58f89ba416399"}},{"code_sha256_prefix":"7df8b90dc252b116","entry":"change_args","repo":"dmitrijsk/attentionhtr","repo_kind":"listed","path":"datasets/process_imgur5k.py","file_url":"https://github.com/dmitrijsk/attentionhtr/blob/HEAD/datasets/process_imgur5k.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7df8b90dc252b116"}},{"code_sha256_prefix":"c7b92e1633b0d2da","entry":"checkImageIsValid","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"create_lmdb_dataset.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/create_lmdb_dataset.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c7b92e1633b0d2da"}},{"code_sha256_prefix":"8a4fd52ddbb2af6b","entry":"create_vitstr","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"modules/vitstr.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/modules/vitstr.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8a4fd52ddbb2af6b"}},{"code_sha256_prefix":"fe2c44cd738c9d3e","entry":"dynamic_lr","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/lr_schedule.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/lr_schedule.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fe2c44cd738c9d3e"}},{"code_sha256_prefix":"ce83d7152f32b55d","entry":"get_args","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ce83d7152f32b55d"}},{"code_sha256_prefix":"98206309c9298631","entry":"get_args","repo":"roatienza/deep-text-recognition-benchmark","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/roatienza/deep-text-recognition-benchmark/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"98206309c9298631"}},{"code_sha256_prefix":"291192e39061a450","entry":"get_device","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"291192e39061a450"}},{"code_sha256_prefix":"edfd158c1fa7f00e","entry":"get_flops","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"predict.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/predict.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"edfd158c1fa7f00e"}},{"code_sha256_prefix":"55becdbd1817adde","entry":"get_image_split","repo":"dmitrijsk/attentionhtr","repo_kind":"listed","path":"datasets/process_imgur5k.py","file_url":"https://github.com/dmitrijsk/attentionhtr/blob/HEAD/datasets/process_imgur5k.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"55becdbd1817adde"}},{"code_sha256_prefix":"d8e30cd1ae314959","entry":"img2text","repo":"roatienza/deep-text-recognition-benchmark","repo_kind":"listed","path":"infer.py","file_url":"https://github.com/roatienza/deep-text-recognition-benchmark/blob/HEAD/infer.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d8e30cd1ae314959"}},{"code_sha256_prefix":"8f99ed40edd07e1e","entry":"isless","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"dataset.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/dataset.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8f99ed40edd07e1e"}},{"code_sha256_prefix":"d6a99ec714147ac1","entry":"linear_warmup_learning_rate","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/lr_schedule.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/lr_schedule.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d6a99ec714147ac1"}},{"code_sha256_prefix":"b2108b927c53418b","entry":"merge","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/model_utils/config.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/model_utils/config.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b2108b927c53418b"}},{"code_sha256_prefix":"3498f707d22460cd","entry":"parse_cli_to_yaml","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/model_utils/config.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/model_utils/config.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3498f707d22460cd"}},{"code_sha256_prefix":"c8d98ae32946f6f0","entry":"parse_yaml","repo":"JoegameZhou/cnnctc","repo_kind":"listed","path":"src/model_utils/config.py","file_url":"https://github.com/JoegameZhou/cnnctc/blob/HEAD/src/model_utils/config.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c8d98ae32946f6f0"}},{"code_sha256_prefix":"e7cc5ea8a6820dc3","entry":"resize","repo":"dmitrijsk/attentionhtr","repo_kind":"listed","path":"datasets/process_imgur5k.py","file_url":"https://github.com/dmitrijsk/attentionhtr/blob/HEAD/datasets/process_imgur5k.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e7cc5ea8a6820dc3"}},{"code_sha256_prefix":"b78e81e855133e45","entry":"tensor2im","repo":"Eom-taeseon/CV_SceneTextRecognition","repo_kind":"listed","path":"dataset.py","file_url":"https://github.com/Eom-taeseon/CV_SceneTextRecognition/blob/HEAD/dataset.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b78e81e855133e45"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}