{"id":"W2910102176","doi":"10.1007/s00138-019-01004-0","title":"Hard negative mining for correlation filters in visual tracking","year":2019,"lang":"en","type":"article","venue":"Machine Vision and Applications","topic":"Video Surveillance and Tracking Methods","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"","keywords":"Computer science; Artificial intelligence; Eye tracking; Tracking (education); Frame (networking); Computer vision; Video tracking; Task (project management); Correlation; Pattern recognition (psychology); Noise (video); Object (grammar); Image (mathematics); Mathematics; Engineering","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":[0.0004058974,0.00008247789,0.0001197373,0.0001093675,0.00009760566,0.00009205126,0.0001515953,0.00003724006,0.000006878157],"category_scores_gemma":[0.00003692028,0.00007371887,0.00003178861,0.0002930096,0.00001457965,0.0002431758,0.00005175077,0.00007485871,0.00001416114],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001373531,"about_ca_system_score_gemma":0.00001333925,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001640364,"about_ca_topic_score_gemma":0.00001343488,"domain_scores_codex":[0.9992831,0.00004585979,0.0001697731,0.0002938008,0.000082534,0.000124924],"domain_scores_gemma":[0.9992455,0.0004322047,0.00006898335,0.0001814782,0.00003507625,0.00003673287],"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":[0.00001311505,0.00007229347,0.05387539,0.00002551578,0.000006003654,2.811124e-7,0.0006701509,0.0009684137,0.002932338,0.02230419,0.00008769071,0.9190446],"study_design_scores_gemma":[0.0008084932,0.0001035824,0.268517,0.00003587271,0.000002938356,0.000003130846,0.00006561168,0.7177649,0.0004787698,0.004853825,0.007199001,0.0001669173],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09865366,0.00006010546,0.8995928,0.000571654,0.00007355133,0.0004798797,0.00000448973,0.00005777293,0.0005061358],"genre_scores_gemma":[0.905338,0.00001091816,0.09413846,0.0002000557,0.00002672493,0.0001061324,0.00001306906,0.00000670081,0.0001599684],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9188777,"threshold_uncertainty_score":0.3006169,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01903159962725122,"score_gpt":0.3370084789500506,"score_spread":0.3179768793227994,"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."}}