{"id":"W4408655814","doi":"10.1016/j.isprsjprs.2025.03.001","title":"3D semantic segmentation: Cluster-based sampling and proximity hashing for novel class discovery","year":2025,"lang":"en","type":"article","venue":"ISPRS Journal of Photogrammetry and Remote Sensing","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; University of Waterloo","funders":"China Scholarship Council; National Natural Science Foundation of China","keywords":"Computer science; Segmentation; Class (philosophy); Hash function; Artificial intelligence; Sampling (signal processing); Cluster (spacecraft); Pattern recognition (psychology); Data mining; Computer vision; Computer security; Programming language","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007410008,0.0001627316,0.0002983101,0.0003002044,0.0002773807,0.0004852331,0.0001618849,0.00007960168,1.033622e-7],"category_scores_gemma":[0.0002755452,0.0001360575,0.0001033443,0.0004104017,0.00008841812,0.001005483,0.0001076993,0.000259774,3.293522e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004688664,"about_ca_system_score_gemma":0.0000886055,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003799407,"about_ca_topic_score_gemma":0.00001195688,"domain_scores_codex":[0.9988489,0.00004873033,0.0004351933,0.0002515785,0.0001795232,0.0002360453],"domain_scores_gemma":[0.9987023,0.0005117461,0.0003234192,0.0001696472,0.0002141135,0.00007871407],"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":[0.00009780344,0.0000295873,0.00008264445,0.0002063958,0.0000424003,0.00001312891,0.0001480131,0.00009674473,0.07171133,0.0001438308,0.00001395321,0.9274142],"study_design_scores_gemma":[0.00254016,0.000409945,0.000301774,0.001699251,0.0001194186,0.0004387855,0.0002953263,0.6128604,0.3684804,0.01015901,0.002332854,0.0003627429],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03383373,0.0005957201,0.9643127,0.0006181867,0.0002926731,0.000276377,0.0000017207,0.00003964535,0.00002926966],"genre_scores_gemma":[0.3803947,0.00007528103,0.6188471,0.0006033859,0.00005458257,6.67082e-8,6.053153e-7,0.000006719089,0.00001753611],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9270514,"threshold_uncertainty_score":0.5548263,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02564280888686539,"score_gpt":0.3178764107897737,"score_spread":0.2922336019029084,"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."}}