{"id":"W4408091964","doi":"10.1117/12.3060400","title":"A comparative study on facial expression recognition using MobileNetV2, VGG-16, ResNet and Swin Transformer","year":2025,"lang":"en","type":"article","venue":"","topic":"Face and Expression Recognition","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Semtech (Canada)","funders":"","keywords":"Residual neural network; Transformer; Artificial intelligence; Computer science; Facial expression recognition; Pattern recognition (psychology); Facial expression; Facial recognition system; Speech recognition; Computer vision; Deep learning; Engineering; Electrical engineering; Voltage","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":[0.0001869561,0.000171248,0.0002212987,0.0002187251,0.000254015,0.0001476919,0.0001918454,0.00007921676,0.0000625477],"category_scores_gemma":[0.00001092367,0.0001357694,0.00004139216,0.0003050837,0.00003151962,0.0005102126,0.00006648669,0.0001510558,0.00004200892],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003986765,"about_ca_system_score_gemma":0.00004007352,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006487854,"about_ca_topic_score_gemma":0.00004868515,"domain_scores_codex":[0.9986924,0.0001605627,0.0002505969,0.0004674781,0.0002265791,0.0002024],"domain_scores_gemma":[0.9994907,0.00009470311,0.00005164184,0.0002224697,0.00007394693,0.00006654496],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00108412,0.00417534,0.004363025,0.0001219943,0.0001953739,0.00003468956,0.02914545,0.0004388813,0.4289286,0.0008751288,0.01900474,0.5116326],"study_design_scores_gemma":[0.007222194,0.002022839,0.01449267,0.00116789,0.00008482335,0.00001183678,0.01514484,0.04397887,0.8985082,0.01243983,0.003776563,0.00114944],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8648342,0.00003144412,0.1223368,0.0001510145,0.0002311858,0.0006614442,0.000006769041,0.0001155269,0.01163158],"genre_scores_gemma":[0.9943435,0.00001266295,0.004928632,0.0003600204,0.00002495545,0.00005853286,0.000009180457,0.000004438422,0.0002580655],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5104831,"threshold_uncertainty_score":0.5536515,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06661826404277414,"score_gpt":0.3328277708738194,"score_spread":0.2662095068310453,"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."}}