{"id":"W3217362767","doi":"10.1109/crv55824.2022.00019","title":"M2A: Motion Aware Attention for Accurate Video Action Recognition","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Human Pose and Action Recognition","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Motion (physics); Focus (optics); Benchmark (surveying); Artificial intelligence; Mechanism (biology); Action (physics); Deep learning; Artificial neural network; Motion estimation; Action recognition; Computer vision; 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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0005342066,0.0003094449,0.0002697162,0.000410818,0.000528333,0.0005189007,0.0005373837,0.0002856471,0.001045785],"category_scores_gemma":[0.00005089564,0.000341306,0.000354686,0.0002467924,0.00001674977,0.00111379,0.0006178014,0.000583392,0.0002439917],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003330062,"about_ca_system_score_gemma":0.0001074813,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005797727,"about_ca_topic_score_gemma":0.00003819249,"domain_scores_codex":[0.9976188,0.0001832386,0.0005089447,0.0009809516,0.0004001117,0.0003079375],"domain_scores_gemma":[0.9983472,0.0001001616,0.0005177145,0.000581959,0.0003655888,0.00008739459],"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":[0.00006685566,0.0003547948,0.00004863274,0.0005827942,0.0001441307,0.00000637017,0.0002220787,0.000855241,0.001321134,0.002279762,0.02007839,0.9740398],"study_design_scores_gemma":[0.003555109,0.0007913974,0.006686394,0.0005403915,0.0004101226,0.00008904043,0.0009010683,0.5232226,0.01603678,0.3704176,0.07406614,0.003283431],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02468926,0.000025065,0.9647914,0.001164344,0.004795018,0.001293601,0.0001335192,0.0008533204,0.002254466],"genre_scores_gemma":[0.92759,0.0005753162,0.03977111,0.001654144,0.002227859,0.004982322,0.0124259,0.0001268089,0.01064651],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9707564,"threshold_uncertainty_score":0.9999039,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1243227273320987,"score_gpt":0.3353635598332738,"score_spread":0.2110408325011751,"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."}}