{"id":"W3197256147","doi":"10.1038/s41598-021-96610-2","title":"Weakly supervised underwater fish segmentation using affinity LCFCN","year":2021,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Water Quality Monitoring Technologies","field":"Environmental Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Montreal Police Service","funders":"","keywords":"Segmentation; Computer science; Artificial intelligence; Convolutional neural network; Annotation; Pattern recognition (psychology); Pixel; Underwater; Geography","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000930118,0.0001427494,0.0001424235,0.00005782416,0.0004206421,0.0004027102,0.0001908197,0.00009079088,0.001006452],"category_scores_gemma":[0.0001515504,0.0001351446,0.00007188766,0.0005905309,0.0003950009,0.0004590694,0.0005266005,0.0001194648,0.0001523436],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002898104,"about_ca_system_score_gemma":0.00004817476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001901508,"about_ca_topic_score_gemma":0.0001140737,"domain_scores_codex":[0.997666,0.00008801473,0.000399485,0.000822154,0.0006587865,0.0003655853],"domain_scores_gemma":[0.9987342,0.00002204094,0.0001436735,0.0009828806,0.00003589833,0.00008132762],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000002025351,0.0001016662,0.08303772,0.00001232266,0.000009056052,0.0004911673,0.0004829821,0.001025126,0.9054149,0.000007472615,0.008140684,0.00127493],"study_design_scores_gemma":[0.00007579312,0.00001089418,0.009924196,0.0000173054,0.00001495136,0.0001804875,0.0004977497,0.0002976545,0.9712645,0.009373347,0.008124111,0.0002189624],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9933794,0.00000923381,0.0005393431,0.0004975508,0.004143534,0.000146178,0.000002247139,0.0002392299,0.001043354],"genre_scores_gemma":[0.9804307,0.000001517681,0.01648382,0.00005126585,0.00003841568,0.000009276928,0.00004564611,0.00001498939,0.002924324],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.07311352,"threshold_uncertainty_score":0.9999068,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05338972691653265,"score_gpt":0.2779017760483843,"score_spread":0.2245120491318516,"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."}}