{"id":"W4409446385","doi":"10.1007/s10044-025-01465-2","title":"Diffusion-based data augmentation and hierarchical CLIP for real estate image annotation","year":2025,"lang":"en","type":"article","venue":"Pattern Analysis and Applications","topic":"Image Retrieval and Classification Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"TransAlta (Canada); Vector Institute; Western University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Annotation; Computer science; Artificial intelligence; Pattern recognition (psychology); Image (mathematics); Diffusion; Computer vision; Physics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002327902,0.00008341125,0.0001309938,0.0002238824,0.0002348959,0.0002135531,0.0003336472,0.00003244829,0.000003404551],"category_scores_gemma":[0.00001510091,0.0000755935,0.00004086568,0.0007638654,0.00006385425,0.0002046547,0.0001343655,0.00004812659,0.00000138531],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001087469,"about_ca_system_score_gemma":0.00002984243,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001339867,"about_ca_topic_score_gemma":0.00005640049,"domain_scores_codex":[0.9990978,0.00003612542,0.0002212499,0.0004454451,0.00009935949,0.0001000447],"domain_scores_gemma":[0.9990218,0.0001734292,0.00009122584,0.0005700263,0.00009566465,0.00004786777],"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":[0.000004019862,0.00006773415,0.006603654,0.00003359411,0.000061056,1.101918e-7,0.00005275284,0.000001832557,0.004620809,0.00651819,0.0001101738,0.9819261],"study_design_scores_gemma":[0.0003926146,0.00002287618,0.09414338,0.000009097727,0.000337981,3.033487e-7,0.00004626553,0.8895953,0.005154093,0.003395628,0.006743893,0.0001585707],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002687257,0.0000195274,0.9933411,0.003071002,0.000007447258,0.0003615868,0.0000978344,0.00008834679,0.0003258689],"genre_scores_gemma":[0.912904,0.0007984687,0.08351682,0.0006195069,0.00003476893,0.0006186851,0.001300351,0.00000691615,0.0002004864],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9817675,"threshold_uncertainty_score":0.3082613,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0209281411788538,"score_gpt":0.3314528959702888,"score_spread":0.310524754791435,"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."}}