{"id":"W4360605113","doi":"10.1109/icnc57223.2023.10074246","title":"Shallow- and Deep- fake Image Manipulation Localization Using Deep Learning","year":2023,"lang":"en","type":"article","venue":"","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Artificial intelligence; Deep learning; Computer science; Task (project management); Code (set theory); Inference; Image (mathematics); Computer vision; Image manipulation; Image editing; Engineering","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":[0.0001576113,0.00008034103,0.00007135112,0.0001621373,0.0001162238,0.0002814341,0.0001075181,0.00004103662,0.000008048867],"category_scores_gemma":[0.0001019942,0.00007995776,0.00001892723,0.0006386455,0.00002766817,0.001128508,0.0001454794,0.00005953094,0.00009641444],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003334633,"about_ca_system_score_gemma":0.000007671161,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002452107,"about_ca_topic_score_gemma":0.0000553499,"domain_scores_codex":[0.9992197,0.00002650973,0.0001339211,0.0002485584,0.000199219,0.0001720791],"domain_scores_gemma":[0.9996633,0.00004205303,0.00004670968,0.0001354864,0.00005670877,0.00005572167],"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":[0.000005599106,0.00001871609,0.01091877,0.00004738096,0.00001581133,0.00003778905,0.001475881,0.04596661,0.005188921,0.02364281,0.00004825325,0.9126335],"study_design_scores_gemma":[0.0001062265,0.00002799588,0.008563454,0.000009701793,0.000002836163,0.00001664963,0.00008380054,0.9865983,0.0007636304,0.003429213,0.0003013057,0.00009695436],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.123184,0.00001137402,0.8728524,0.00004738772,0.0002960133,0.00008061886,5.867444e-8,0.0005439512,0.002984161],"genre_scores_gemma":[0.9793434,0.0000074245,0.02027262,0.00005279215,0.00004284721,0.000003200481,0.000007644196,0.00001100578,0.0002590337],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9406316,"threshold_uncertainty_score":0.3260583,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02041804617560285,"score_gpt":0.2444755988105637,"score_spread":0.2240575526349609,"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."}}