{"id":"W4416639837","doi":"10.1061/jtepbs.teeng-9214","title":"Evaluating Semantic Segmentation–Based Scene Descriptions for Multilane Rural Highway Point Clouds against Ground Truth","year":2025,"lang":"en","type":"article","venue":"Journal of Transportation Engineering Part A Systems","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Segmentation; Ground truth; Point cloud; Transformer; Point (geometry); Robustness (evolution); Consistency (knowledge bases); Intersection (aeronautics); Visual reasoning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001125034,0.001406668,0.0004823598,0.002597437,0.0003271399,0.00121453,0.001309682,0.001034437,0.001761364],"category_scores_gemma":[0.004127522,0.0003506423,0.001190856,0.001353843,0.0006681083,0.002278177,0.0009139925,0.0006888837,0.001452681],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001761352,"about_ca_system_score_gemma":0.001030693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02036485,"about_ca_topic_score_gemma":0.02887975,"domain_scores_codex":[0.9990628,0.0001951538,0.00005526258,0.0003215739,0.0002807343,0.00008442593],"domain_scores_gemma":[0.9983177,0.0008371614,0.0001727178,0.000204726,0.0003896516,0.00007806399],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001071271,0.0003985589,0.01656204,0.001061022,0.000333544,0.0008233578,0.001138333,0.5874559,0.03744003,0.0048977,0.01165084,0.3371674],"study_design_scores_gemma":[0.00002194325,0.00009219046,0.002847399,0.00003568813,0.00003058786,0.0001016781,0.0004780519,0.9824836,0.01036773,0.001494411,0.002022266,0.00002435461],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5705258,0.0008381039,0.3872009,0.0005220269,0.0002022841,0.0006911097,0.0129379,0.02046327,0.006618703],"genre_scores_gemma":[0.8274877,0.0002490609,0.1440627,0.0001495527,0.00002645021,0.0002012791,0.02564267,0.0004852545,0.001695324],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02036485,"threshold_uncertainty_score":0.04049265,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01864355470874573,"score_gpt":0.2687108066678812,"score_spread":0.2500672519591355,"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."}}