{"id":"W4386243184","doi":"10.1109/crv60082.2023.00024","title":"Transformer-Based Human Action Recognition with Dynamic Feature Selection","year":2023,"lang":"en","type":"article","venue":"","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Discriminative model; Action recognition; Artificial intelligence; Transformer; Encoder; Leverage (statistics); Machine learning; Pattern recognition (psychology); Human motion; Feature selection; Computer vision; Motion (physics); Engineering","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":[0.0001281934,0.0001245316,0.00008916813,0.0003356245,0.0003406555,0.0001455433,0.0001257455,0.00008708634,0.0001540769],"category_scores_gemma":[0.000003030713,0.0001049862,0.00005274182,0.001023276,0.00001722499,0.0007810235,0.000005179472,0.0001708545,0.0005525096],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006754497,"about_ca_system_score_gemma":0.00004179001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001942574,"about_ca_topic_score_gemma":0.000438288,"domain_scores_codex":[0.9990892,0.0000413525,0.0001126194,0.000311091,0.0002260083,0.0002197388],"domain_scores_gemma":[0.9996319,0.0000236444,0.00005328077,0.0001306986,0.0001022786,0.00005818504],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00008300737,0.0002629458,0.0003154719,0.0001241577,0.00007510043,0.00002412073,0.0003747978,0.0004806544,0.1404276,0.001810847,0.0108708,0.8451505],"study_design_scores_gemma":[0.004923579,0.002419536,0.03849636,0.0002779214,0.0001074753,0.0001781935,0.0004549541,0.4484962,0.4643598,0.0204212,0.01810797,0.001756881],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4098288,0.000002343274,0.5785357,0.001832686,0.0002138849,0.0003015279,0.000004519849,0.002120172,0.007160396],"genre_scores_gemma":[0.9922736,0.000008351516,0.004504773,0.0003152272,0.00005757614,0.00005636223,0.0002023207,0.00001469054,0.002567041],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8433936,"threshold_uncertainty_score":0.7101577,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03423737450805013,"score_gpt":0.285123491204596,"score_spread":0.2508861166965459,"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."}}