{"id":"W4210926750","doi":"10.1101/2022.02.07.479461","title":"CameraTrapDetectoR: Automatically detect, classify, and count animals in camera trap images using artificial intelligence","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Wildlife Ecology and Conservation","field":"Environmental Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan; National Capital Commission","funders":"","keywords":"Camera trap; Artificial intelligence; Computer science; Bounding overwatch; Computer vision; Minimum bounding box; Deep learning; Object (grammar); Wildlife; Pattern recognition (psychology); Class (philosophy); Image (mathematics); Ecology; Biology","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001032371,0.0004698612,0.0005119029,0.0001970162,0.0003292548,0.00017697,0.0004160939,0.0004215782,0.00109381],"category_scores_gemma":[0.0002098421,0.0005481842,0.00008656775,0.00058099,0.0004703397,0.000220135,0.0007962006,0.001067199,0.00005120333],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007987804,"about_ca_system_score_gemma":0.0002764248,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005257491,"about_ca_topic_score_gemma":0.0001369477,"domain_scores_codex":[0.9968499,0.0003368683,0.0007400594,0.0010655,0.0004237826,0.0005839115],"domain_scores_gemma":[0.9986287,0.0001594044,0.0003604027,0.0006267753,0.00003883402,0.0001858471],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0001164346,0.0002621925,0.3297844,0.0002079253,0.0000902321,0.0002834429,0.00007536596,0.003518132,0.6645024,0.0009546441,0.0001159444,0.00008886118],"study_design_scores_gemma":[0.0001373734,0.0001172765,0.9447244,0.00009693209,0.00008100089,1.761381e-7,0.0000197015,0.01952623,0.03397486,0.0001420589,0.000319755,0.0008602174],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9946445,0.0003238842,0.003278309,0.0003236491,0.000476383,0.0006778142,0.00008552221,0.0001755589,0.00001438975],"genre_scores_gemma":[0.9886698,0.0001046333,0.01052975,0.0003461139,0.00009538409,0.0001894537,1.814996e-7,0.00006330832,0.00000135193],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6305276,"threshold_uncertainty_score":0.9998193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02270265562780522,"score_gpt":0.2401486393303277,"score_spread":0.2174459837025225,"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."}}