{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005122295,0.001743622,0.002212402,0.003150873,0.001041409,0.003982675,0.004861739,0.003374209,0.001939516],"category_scores_gemma":[0.01945505,0.0007347059,0.0007489492,0.002609136,0.001870965,0.009366632,0.002592236,0.002831308,0.003990866],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001880074,"about_ca_system_score_gemma":0.001234178,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008497889,"about_ca_topic_score_gemma":0.008215713,"domain_scores_codex":[0.9941856,0.001263647,0.0005433955,0.001576051,0.002042793,0.0003885169],"domain_scores_gemma":[0.9901277,0.004510224,0.0008069379,0.001636492,0.00261023,0.0003082096],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004041301,0.0003701636,0.005848932,0.001313994,0.0001635188,0.0004692797,0.0002663193,0.06556796,0.01976873,0.007267703,0.06018364,0.8383756],"study_design_scores_gemma":[0.00004641586,0.0001689433,0.004846439,0.000247902,0.00006981562,0.0006314529,0.000661554,0.8874001,0.03795841,0.04523341,0.02265646,0.00007920205],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1864099,0.02280878,0.7193404,0.01641807,0.00183489,0.0005146423,0.01172723,0.02314874,0.01779743],"genre_scores_gemma":[0.5798803,0.006161723,0.3780626,0.0024924,0.0007753985,0.00036175,0.02025739,0.000868405,0.01114003],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008497889,"threshold_uncertainty_score":0.0270896,"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."}}