{"id":"W3097711533","doi":"10.1016/j.patcog.2020.107737","title":"Deep feature augmentation for occluded image classification","year":2020,"lang":"en","type":"article","venue":"Pattern Recognition","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":74,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Artificial intelligence; Computer science; Pattern recognition (psychology); Convolutional neural network; Feature (linguistics); Image (mathematics); Set (abstract data type); Deep learning; Feature extraction; Contextual image classification; Feature vector; Computer vision","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"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":[0.00005468259,0.000107988,0.00008787597,0.00003402508,0.0001294976,0.00009582693,0.0002694619,0.0000515478,0.00002041392],"category_scores_gemma":[0.00003860327,0.0001150189,0.00005368412,0.0002717889,0.00001598142,0.000600541,0.00004368394,0.00009314468,0.0002496933],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003612609,"about_ca_system_score_gemma":0.000009766566,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":9.413529e-7,"about_ca_topic_score_gemma":0.000002488249,"domain_scores_codex":[0.9991179,0.0000342234,0.0001603678,0.0003942943,0.0001292242,0.0001639823],"domain_scores_gemma":[0.9993587,0.00009213018,0.0001378408,0.0001932615,0.000130749,0.00008728679],"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":[0.00001732616,0.00003875079,0.0001491639,0.00003975994,0.000009874262,0.000001717185,0.0004113759,0.00005105593,0.0354116,0.0003303569,0.003823678,0.9597154],"study_design_scores_gemma":[0.001023369,0.0001470713,0.005270925,0.00002116919,0.00002615715,0.0000145178,0.0000858729,0.947752,0.02965145,0.01271555,0.002927541,0.0003643612],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003333488,0.0000183759,0.9819051,0.01340131,0.0001110101,0.0006485594,0.00002152594,0.0002527839,0.0003078576],"genre_scores_gemma":[0.7340843,0.00002332613,0.2597837,0.004340901,0.0003869782,0.0006221149,0.0007080576,0.00002316254,0.00002749348],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.959351,"threshold_uncertainty_score":0.4690333,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06730579002303476,"score_gpt":0.3001896617219861,"score_spread":0.2328838716989513,"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."}}