{"id":"W4407385589","doi":"10.3390/jimaging11020052","title":"SAVE: Self-Attention on Visual Embedding for Zero-Shot Generic Object Counting","year":2025,"lang":"en","type":"article","venue":"Journal of Imaging","topic":"Video Surveillance and Tracking Methods","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University; Université TÉLUQ; Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Artificial intelligence; Object (grammar); Embedding; Shot (pellet); Computer vision; Feature (linguistics); Class (philosophy); Pattern recognition (psychology); Zero (linguistics); Matching (statistics); Mathematics; Statistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007139082,0.001802978,0.001467467,0.001396707,0.0003845679,0.001193427,0.002663707,0.001222728,0.004819433],"category_scores_gemma":[0.002450347,0.0005073098,0.0009103497,0.001015776,0.0007899891,0.002798705,0.002650186,0.001299466,0.002051555],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008027186,"about_ca_system_score_gemma":0.0008778643,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0065347,"about_ca_topic_score_gemma":0.008498231,"domain_scores_codex":[0.999255,0.0001076367,0.00003040791,0.0002988794,0.0002023897,0.0001056741],"domain_scores_gemma":[0.9992058,0.0002077948,0.0000800456,0.0002685728,0.0001802991,0.0000575479],"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":[0.0003280604,0.0002565313,0.00116117,0.0002052129,0.00009303377,0.0001037677,0.0001166602,0.02749866,0.02559268,0.004581484,0.01217919,0.9278835],"study_design_scores_gemma":[0.00003203755,0.0002061172,0.001059648,0.00003213529,0.00004478416,0.0002130831,0.00006813747,0.957361,0.0255222,0.009703268,0.005725354,0.00003221741],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04426664,0.0008858927,0.9357406,0.0001687215,0.0002378497,0.0001785069,0.0006298128,0.01531703,0.002574882],"genre_scores_gemma":[0.5149858,0.0007030443,0.4645694,0.0005248217,0.0001601062,0.0002187281,0.004197286,0.00132081,0.01331995],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.0065347,"threshold_uncertainty_score":0.01612258,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02050528559219349,"score_gpt":0.3571574866703086,"score_spread":0.3366522010781151,"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."}}