{"id":"W2888516611","doi":"10.1109/access.2018.2866089","title":"Sparsity-Based Image Inpainting Detection via Canonical Correlation Analysis With Low-Rank Constraints","year":2018,"lang":"en","type":"article","venue":"IEEE Access","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"National Natural Science Foundation of China","keywords":"Inpainting; Artificial intelligence; Computer science; Pattern recognition (psychology); Image (mathematics); JPEG; Feature (linguistics); Computer vision; Mathematics","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":[0.0003295079,0.000163508,0.0002111601,0.000414485,0.0001877525,0.0005627437,0.0005885776,0.00008649394,0.0000258872],"category_scores_gemma":[0.0000913328,0.0001495734,0.00008840017,0.002154771,0.0003813821,0.00182596,0.00008129895,0.0001665302,0.00008405292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001157049,"about_ca_system_score_gemma":0.00009675425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000247283,"about_ca_topic_score_gemma":0.002648022,"domain_scores_codex":[0.998458,0.00007425645,0.0002569105,0.0004891488,0.0004154576,0.0003061741],"domain_scores_gemma":[0.9987229,0.0001327421,0.0002108757,0.0004841205,0.0003265422,0.0001228791],"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":[0.0003354237,0.0003975026,0.06895056,0.00008969891,0.0008293115,0.0001677914,0.0007880144,0.01011146,0.02804821,0.0003242542,0.0002248716,0.8897329],"study_design_scores_gemma":[0.0008292705,0.0003360981,0.04898901,0.00005077457,0.0001924512,0.00003259793,0.00001010611,0.6955759,0.2529291,0.0005459942,0.00009506856,0.000413645],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3736274,8.724523e-7,0.6242182,0.00005450076,0.0006412941,0.0001150197,0.000001323871,0.0001720841,0.001169323],"genre_scores_gemma":[0.9907155,1.608238e-7,0.008876362,0.0001928058,0.000168654,0.00001248064,0.000004703023,0.00001095615,0.00001835715],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8893192,"threshold_uncertainty_score":0.6099424,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009953577584668239,"score_gpt":0.2505138867084685,"score_spread":0.2405603091238003,"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."}}