{"id":"W2602628422","doi":"10.1109/crv.2017.20","title":"Multi-path Region-Based Convolutional Neural Network for Accurate Detection of Unconstrained \"Hard Faces\"","year":2017,"lang":"en","type":"article","venue":"","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Nvidia","keywords":"Computer science; Artificial intelligence; Convolutional neural network; Pattern recognition (psychology); Pixel; Face detection; Classifier (UML); Face (sociological concept); Deep learning; Facial recognition system; Feature (linguistics); Convolution (computer science); Path (computing); Feature extraction; Artificial neural network","routes":{"ca_aff":true,"ca_fund":true,"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":[0.000205623,0.000102469,0.0001331757,0.00003709887,0.0004210541,0.0001182225,0.0004581381,0.00007134912,0.00001132682],"category_scores_gemma":[0.0001277511,0.00008590384,0.00009947355,0.00005486291,0.00008869951,0.0004719394,0.00006464376,0.00006207565,0.00000964723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001566387,"about_ca_system_score_gemma":0.00007156601,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005859349,"about_ca_topic_score_gemma":0.00004557603,"domain_scores_codex":[0.9991667,0.00004008043,0.0001989052,0.000255773,0.0001315206,0.0002070254],"domain_scores_gemma":[0.9989535,0.0001368023,0.0002540541,0.0004036518,0.0001888331,0.00006311832],"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":[0.0009445521,0.0007946681,0.02368914,0.0003744195,0.00019838,0.00003096501,0.0004329564,0.07674621,0.3268424,0.01314295,0.03543625,0.5213671],"study_design_scores_gemma":[0.001207909,0.00009047524,0.009943667,0.00004005158,0.000005902159,0.000003794624,0.00001314867,0.9663193,0.02098802,0.0007592784,0.0004985897,0.0001298398],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03443921,0.00001594179,0.9635431,0.0008054288,0.0006693993,0.0002713124,0.0000112486,0.00008576783,0.0001586573],"genre_scores_gemma":[0.9133367,0.000002019025,0.08608709,0.0002098843,0.00007297602,0.00003992502,0.000009504738,0.000004496174,0.0002373871],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8895731,"threshold_uncertainty_score":0.3503057,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06537923013406781,"score_gpt":0.2879696959416588,"score_spread":0.222590465807591,"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."}}