{"id":"W3127551618","doi":"10.15353/jcvis.v6i1.3543","title":"Challenges of Deep Learning-based Text Detection in the Wild","year":2021,"lang":"en","type":"article","venue":"Journal of Computational Vision and Imaging Systems","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"ATS Automation Tooling Systems (Canada); University of Waterloo","funders":"","keywords":"Benchmark (surveying); Computer science; Artificial intelligence; Distortion (music); Deep learning; Perspective (graphical); Machine learning; Text detection; Function (biology); Pattern recognition (psychology); Image (mathematics)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001101948,0.00007195642,0.0001755635,0.0002354227,0.00005965553,0.0001372776,0.0001936177,0.00002453843,0.00000194426],"category_scores_gemma":[0.0001145652,0.00005119247,0.00006101854,0.0002422857,0.00003025332,0.0003346094,0.00002879991,0.0001931251,0.000001178256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002199832,"about_ca_system_score_gemma":0.00005826175,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004727756,"about_ca_topic_score_gemma":0.000002775521,"domain_scores_codex":[0.9985575,0.0004069587,0.0004384469,0.0001008111,0.0004184681,0.00007778403],"domain_scores_gemma":[0.9985515,0.0004528792,0.0003669959,0.00008466052,0.0005123889,0.00003153274],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001601557,0.0001756837,0.0009732041,0.0001293774,0.00001861687,0.00007498001,0.001601926,0.02520248,0.001159105,0.004389409,0.000103687,0.9661555],"study_design_scores_gemma":[0.0007981189,0.0002520964,0.03274871,0.0005784826,0.000008615631,0.001445468,0.00140927,0.9527334,0.001712336,0.006539374,0.001651163,0.0001229453],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01732869,0.004282322,0.9753892,0.002385383,0.0001824217,0.00006532639,2.578555e-7,0.00001862498,0.0003477473],"genre_scores_gemma":[0.9935632,0.00009392055,0.006154005,0.0001338854,0.00004496395,0.000001871148,5.905978e-7,0.000003902671,0.000003656701],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9762345,"threshold_uncertainty_score":0.2087568,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01399929453120768,"score_gpt":0.2751410523217244,"score_spread":0.2611417577905167,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}