{"id":"W4205196999","doi":"10.18280/ts.380611","title":"Non-Global Privacy Protection Facing Sensitive Areas in Face Images","year":2021,"lang":"en","type":"article","venue":"Traitement du signal","topic":"Face recognition and analysis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Science Foundation of Heilongjiang Province; National Natural Science Foundation of China","keywords":"Landmark; Differential privacy; Computer science; Face (sociological concept); Image (mathematics); Privacy protection; Position (finance); Noise (video); Computer vision; Similarity (geometry); Digital watermarking; Data mining; Artificial intelligence; Computer security; Business","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001916176,0.0001273554,0.0001597149,0.00007966894,0.00008454736,0.0001640677,0.0001591426,0.00003827102,0.0001025482],"category_scores_gemma":[0.00001992097,0.0001281524,0.00009221501,0.0006600764,0.00001912062,0.0003759174,0.0001081245,0.0001080417,0.00008416441],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001162019,"about_ca_system_score_gemma":0.00007588582,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006582547,"about_ca_topic_score_gemma":0.00004307925,"domain_scores_codex":[0.9988053,0.00008885653,0.0002197606,0.0003774692,0.0002739167,0.0002346276],"domain_scores_gemma":[0.9995784,0.00001790648,0.00006230499,0.0001550706,0.0001137958,0.00007257775],"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":[0.00006220676,0.001382762,0.0112826,0.0001647586,0.000377667,0.001339725,0.006977474,0.01475975,0.2523903,0.002680597,0.001147742,0.7074345],"study_design_scores_gemma":[0.002777042,0.0001795235,0.09088643,0.0003030234,0.0000544759,0.0001637662,0.001941939,0.5099412,0.3897836,0.001876871,0.001214109,0.0008779732],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2248641,0.00002087617,0.7719722,0.001662984,0.0000609939,0.0001789293,0.00001007251,0.00007208236,0.001157759],"genre_scores_gemma":[0.9949368,0.000008269902,0.004584013,0.0002697459,0.00003588428,0.0000220833,0.0000113078,0.000003752819,0.0001281516],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7700727,"threshold_uncertainty_score":0.5225902,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01775190793796452,"score_gpt":0.2428528727575442,"score_spread":0.2251009648195797,"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."}}