{"id":"W4221166146","doi":"10.1016/j.jvcir.2023.103800","title":"TransCAM: Transformer attention-based CAM refinement for Weakly supervised semantic segmentation","year":2023,"lang":"en","type":"article","venue":"Journal of Visual Communication and Image Representation","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":85,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto; University of New Brunswick","funders":"","keywords":"Computer science; Artificial intelligence; Transformer; Segmentation; Convolutional neural network; Discriminative model; Pattern recognition (psychology); Pixel; Pascal (unit); Voltage","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":[0.0005741373,0.0001152762,0.0001776371,0.0002585741,0.000282292,0.0001407475,0.000397522,0.00003922619,0.000005744427],"category_scores_gemma":[0.00005767221,0.0001094622,0.0001257403,0.0006742979,0.00005835291,0.001033093,0.0000277669,0.0001277491,0.000005696867],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004498115,"about_ca_system_score_gemma":0.00004953121,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007553774,"about_ca_topic_score_gemma":0.000007372734,"domain_scores_codex":[0.9985999,0.0001666207,0.0005921722,0.0001815694,0.0003007215,0.0001590456],"domain_scores_gemma":[0.9983785,0.0004297051,0.0003629237,0.0003510407,0.0003977856,0.00008003117],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002020868,0.0004648291,0.001161647,0.0001710258,0.0001029703,0.000002644282,0.001776731,0.002698581,0.7782105,0.00457035,0.005481778,0.2051569],"study_design_scores_gemma":[0.005552096,0.0008460499,0.02904127,0.0001958808,0.0001526941,0.00004535746,0.001719754,0.8727619,0.08064754,0.005289739,0.003280358,0.0004673614],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0867136,0.0001190559,0.9028779,0.009402143,0.0001066911,0.0005736342,0.000005455351,0.00007660734,0.000124886],"genre_scores_gemma":[0.8770842,0.001070744,0.1211311,0.0002556654,0.00009046634,0.00011662,0.000125433,0.00001879743,0.0001069461],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8700633,"threshold_uncertainty_score":0.446374,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04025519472113439,"score_gpt":0.3715744347563467,"score_spread":0.3313192400352123,"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."}}