{"id":"W4407360190","doi":"10.36227/techrxiv.173933251.15884096/v1","title":"Global-Local Image Perceptual Score (GLIPS): Evaluating Photorealistic Quality of AI-Generated Images","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Visual Attention and Saliency Detection","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Artificial intelligence; Computer vision; Perception; Computer science; Image (mathematics); Image quality; Quality (philosophy); Psychology; Physics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001300287,0.0006148348,0.0003290542,0.001660861,0.000203107,0.001038068,0.0004796425,0.0005018605,0.002983047],"category_scores_gemma":[0.008864935,0.0001316226,0.0003807805,0.0005903696,0.0005749037,0.00128207,0.001043859,0.0004639438,0.0005884169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000481272,"about_ca_system_score_gemma":0.0002706848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000793046,"about_ca_topic_score_gemma":0.0008810418,"domain_scores_codex":[0.9992792,0.0001332463,0.00003963713,0.0001024288,0.0003995608,0.00004598172],"domain_scores_gemma":[0.9969531,0.001212627,0.0004340953,0.0003208495,0.0008539349,0.0002253348],"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":[0.001544817,0.0002487263,0.02662224,0.001438958,0.000346951,0.000383624,0.0008890022,0.0374732,0.3272713,0.01015356,0.006177709,0.5874498],"study_design_scores_gemma":[0.0001919934,0.004259855,0.1717257,0.0002542222,0.0004547517,0.003166483,0.001329119,0.4514094,0.3244683,0.0220785,0.02029155,0.0003701423],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4842349,0.001659105,0.4882956,0.000310904,0.0002374285,0.0005445614,0.001057293,0.002642548,0.02101773],"genre_scores_gemma":[0.9001596,0.0003388129,0.09655179,0.00009278164,0.00004860295,0.0001151018,0.0006955159,0.0002849988,0.001712698],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002983047,"threshold_uncertainty_score":0.009979248,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0856993687730757,"score_gpt":0.4176776980808296,"score_spread":0.3319783293077538,"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."}}