{"id":"W2992457155","doi":"10.1109/iccvw.2019.00189","title":"Interpretable Spatio-Temporal Attention for Video Action Recognition","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":103,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Discriminative model; Interpretability; Computer science; Artificial intelligence; Pattern recognition (psychology); Salient; Action recognition; Set (abstract data type); Feature extraction; Minimum bounding box; Feature (linguistics); Bounding overwatch; Machine learning; Computer vision; Image (mathematics); Class (philosophy)","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"],"consensus_categories":[],"category_scores_codex":[0.0003754118,0.0002692424,0.0002873477,0.0003102254,0.0001347702,0.0005518663,0.000446022,0.0002799208,0.0004030607],"category_scores_gemma":[0.00004123726,0.0002773356,0.0002989162,0.000120009,0.00001512747,0.00109609,0.0003652105,0.0003628285,0.0007695999],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001625249,"about_ca_system_score_gemma":0.0001315152,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001165051,"about_ca_topic_score_gemma":0.0001078419,"domain_scores_codex":[0.998166,0.00007704063,0.000448493,0.0007892285,0.0002534523,0.0002657895],"domain_scores_gemma":[0.998468,0.0000828473,0.0004316671,0.0005734446,0.0003740078,0.00007002139],"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":[0.0002570066,0.0005935854,0.001016884,0.001780332,0.0003847236,0.00000558892,0.0005199497,0.0007162688,0.001971221,0.004124396,0.1060887,0.8825414],"study_design_scores_gemma":[0.002037143,0.0005147881,0.001400567,0.001174172,0.0001847206,0.00003667949,0.0001418355,0.6842327,0.01672585,0.2227826,0.06893045,0.001838486],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04419469,0.00002131901,0.9426168,0.0004346902,0.00455902,0.00121378,0.00004914531,0.0004960755,0.006414469],"genre_scores_gemma":[0.9313895,0.00004513345,0.05778766,0.0008243062,0.0006351734,0.0005387506,0.002537594,0.00003826587,0.006203697],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8871948,"threshold_uncertainty_score":0.9999679,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0665918228806321,"score_gpt":0.3027020723865421,"score_spread":0.23611024950591,"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."}}