{"id":"W4387814628","doi":"10.1007/s10344-023-01742-7","title":"The DeepFaune initiative: a collaborative effort towards the automatic identification of European fauna in camera trap images","year":2023,"lang":"en","type":"article","venue":"European Journal of Wildlife Research","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Office Français de la Biodiversité; Région Normandie; Région Auvergne-Rhône-Alpes; Centre National de la Recherche Scientifique; Région Occitanie Pyrénées-Méditerranée; Trent University; Nottingham Trent University","keywords":"Wildlife; Camera trap; Computer science; Citizen science; Milestone; Identification (biology); Fauna; Software; Artificial intelligence; Data science; Cartography; Ecology; Geography; Biology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.01501632,0.0001066168,0.0001573242,0.0001484895,0.0003934611,0.0001550838,0.0009087687,0.00001342411,0.001002973],"category_scores_gemma":[0.001079103,0.00006176878,0.00008272701,0.002039497,0.0008532878,0.0002211503,0.000318269,0.000474449,0.0009680427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002657858,"about_ca_system_score_gemma":0.00006141603,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001775631,"about_ca_topic_score_gemma":0.00002590355,"domain_scores_codex":[0.9943925,0.003144349,0.0007523819,0.0001555445,0.001182767,0.0003724585],"domain_scores_gemma":[0.9986023,0.0004128753,0.0003810292,0.0002898465,0.0002107514,0.0001031487],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.0004949257,0.0007930017,0.07019947,0.0001396007,0.000279796,0.001511126,0.04507025,0.0009065975,0.04242381,0.00143447,0.6209283,0.2158187],"study_design_scores_gemma":[0.0004577865,0.0001871016,0.9585029,0.00005325554,0.000007124531,0.00002291336,0.01864642,0.0002023916,0.0009597276,0.00006372893,0.02082236,0.00007431376],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9463793,0.0002239504,0.00004646637,0.009511357,0.0001323007,0.0002982685,0.00004189904,0.00001805818,0.04334842],"genre_scores_gemma":[0.9982102,0.001148939,0.00001391435,0.0001090591,0.00006713705,0.000004584455,0.000006594255,0.0000219068,0.0004176492],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8883034,"threshold_uncertainty_score":0.9999102,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0896938462887525,"score_gpt":0.3554394021795951,"score_spread":0.2657455558908426,"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."}}