{"id":"W4403531947","doi":"10.5194/isprs-annals-x-4-2024-141-2024","title":"Semantic Segmentation Uncertainty Assessment of Different U-net Architectures for Extracting Building Footprints","year":2024,"lang":"en","type":"article","venue":"ISPRS annals of the photogrammetry, remote sensing and spatial information sciences","topic":"BIM and Construction Integration","field":"Engineering","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Segmentation; Computer science; Artificial intelligence; Natural language processing","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":[0.0004434464,0.0001056102,0.0001390522,0.0002194964,0.0001527261,0.0001531779,0.00007293474,0.00004295333,0.000002459662],"category_scores_gemma":[0.00007280063,0.00007112074,0.00009064803,0.0002892357,0.0001354214,0.0000958807,0.00002122552,0.00009767515,1.203869e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001542649,"about_ca_system_score_gemma":0.000027564,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006583196,"about_ca_topic_score_gemma":0.001173874,"domain_scores_codex":[0.9990956,0.00002970165,0.0003853443,0.00009089864,0.0002585387,0.0001399473],"domain_scores_gemma":[0.9994882,0.0001581277,0.0001350234,0.00008497906,0.0001043804,0.00002924884],"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":[0.000004587944,0.000002285185,0.00009338698,0.0001887215,0.00002347793,3.388195e-8,0.0004136065,0.02592373,0.01309654,0.00009345417,0.0000212229,0.9601389],"study_design_scores_gemma":[0.00006482074,0.00004448391,0.001418556,0.0002124364,0.0000140427,0.000009001637,0.0004756688,0.9071048,0.08829436,0.002074688,0.0002146132,0.00007250337],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4296544,0.00004967679,0.5691277,0.0001531628,0.0005161787,0.000169236,0.00001071145,0.00003604672,0.0002828196],"genre_scores_gemma":[0.9964688,0.00003408572,0.00341377,0.00003631563,0.00003452289,5.935041e-7,0.000004115661,0.000004481992,0.000003267189],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9600664,"threshold_uncertainty_score":0.9951866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03155067032171511,"score_gpt":0.3245719354545922,"score_spread":0.2930212651328771,"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."}}