{"id":"W3143614868","doi":"10.1109/iccv48922.2021.01371","title":"Labels4Free: Unsupervised Segmentation using StyleGAN","year":2021,"lang":"en","type":"preprint","venue":"2021 IEEE/CVF International Conference on Computer Vision (ICCV)","topic":"Generative Adversarial Networks and Image Synthesis","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"King Abdullah University of Science and Technology","keywords":"Segmentation; Computer science; Artificial intelligence; Segmentation-based object categorization; Scale-space segmentation; Pattern recognition (psychology); Object (grammar); Image segmentation; Generator (circuit theory); Computer vision; Power (physics)","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","scholarly_communication","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0004537742,0.0007659667,0.0007243619,0.0004783559,0.0002852677,0.003159683,0.002779376,0.000408357,0.001144473],"category_scores_gemma":[0.00004217077,0.0007739912,0.0004248006,0.0004123508,0.00008317719,0.00102711,0.0023955,0.0009746671,0.0001378559],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005004277,"about_ca_system_score_gemma":0.0006504015,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003128707,"about_ca_topic_score_gemma":0.00008728388,"domain_scores_codex":[0.9945906,0.0005876534,0.000931406,0.001930525,0.001425758,0.0005339924],"domain_scores_gemma":[0.9960679,0.0001874408,0.0005780868,0.001360045,0.001540899,0.00026558],"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":[0.000165339,0.001517229,0.0003611764,0.0001810092,0.001441319,0.000779175,0.002459436,0.4040166,0.07734988,0.0226273,0.01764862,0.4714529],"study_design_scores_gemma":[0.000682727,0.0001756439,0.0002177961,0.000822629,0.00003578321,0.0000220383,0.0001025695,0.9868928,0.008312802,0.001148395,0.0008146843,0.000772169],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03205578,0.0001013225,0.9490695,0.003268204,0.01316751,0.0004928585,0.00006908058,0.0001459655,0.001629772],"genre_scores_gemma":[0.7179021,0.0002987203,0.2768688,0.001434921,0.00233765,0.00004731613,0.000401036,0.0000575493,0.0006519419],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6858463,"threshold_uncertainty_score":0.9997686,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06634207614877823,"score_gpt":0.3233066719123797,"score_spread":0.2569645957636014,"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."}}