{"id":{"repo_id":"nus","oai_identifier":"oai:scholarbank.nus.edu.sg:10635/198469"},"canonical_url":"https://search.dev.ndltd.org/etd/nus/oai:scholarbank.nus.edu.sg:10635/198469","repository":{"repo_id":"nus","name":"National University of Singapore","base_url":"https://scholarbank.nus.edu.sg/oai/request"},"display":{"title":"DEEP LEARNING FOR FOR LARGE-SCALE FACIAL EXPRESSION RECOGNITION IN THE WILD","abstract":"Human facial expression recognition (FER) is a challenging task due to ambiguous facial expressions and highly skewed, imbalanced datasets. Even with recent advances in deep learning, these obstacles still remain. Facial expressions are commonly assigned discrete emotions, as per the categorical model of affect. However, continuous dimensional models of affect, such as those using valence/arousal, provide significant advantages over categorical models in terms of representing complex human emotional states. In this thesis, a novel method to derive dimensional annotations from typical categorical affect datasets will be proposed. The thesis further shows that dimensional and categorical emotion spaces can be jointly learnt via label fusion in order to obtain better feature representations. Experiments on in-the-wild AffectNet as well as a collection of laboratory based FER datasets suggests that jointly learning both dimensional and categorical models of affect significantly reduce redundancy and computational load whilst improving generalization performance, outperforming state-of-the-art methods for dimensional AffectNet with 0.33 and 0.30 RMSEs.","abstract_html":"Human facial expression recognition (FER) is a challenging task due to ambiguous facial expressions and highly skewed, imbalanced datasets. Even with recent advances in deep learning, these obstacles still remain. Facial expressions are commonly assigned discrete emotions, as per the categorical model of affect. However, continuous dimensional models of affect, such as those using valence/arousal, provide significant advantages over categorical models in terms of representing complex human emotional states. In this thesis, a novel method to derive dimensional annotations from typical categorical affect datasets will be proposed. The thesis further shows that dimensional and categorical emotion spaces can be jointly learnt via label fusion in order to obtain better feature representations. Experiments on in-the-wild AffectNet as well as a collection of laboratory based FER datasets suggests that jointly learning both dimensional and categorical models of affect significantly reduce redundancy and computational load whilst improving generalization performance, outperforming state-of-the-art methods for dimensional AffectNet with 0.33 and 0.30 RMSEs.","abstract_has_math":false,"creators":["NEO YUAN RONG DEXTER"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-07-28","date_published":"2021-07-28","updated_at":"2026-07-24T03:31:51Z","subjects":["Emotions, Deep Learning, Facial Expression Recognition"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":null,"outbound_label":null,"outbound_source":null},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["NEO YUAN RONG DEXTER"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.issued","label":"Date","values":["2021-07-28"]},{"key":"dc:relation.isreferencedby","label":"Dc Relation Isreferencedby","values":["https://scholarbank.nus.edu.sg/handle/10635/198469"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Emotions, Deep Learning, Facial Expression Recognition"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://scholarbank.nus.edu.sg/bitstreams/7c474e6f-ee19-4abd-86b5-967dc1af99d1/download"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Human facial expression recognition (FER) is a challenging task due to ambiguous facial expressions and highly skewed, imbalanced datasets. Even with recent advances in deep learning, these obstacles still remain. Facial expressions are commonly assigned discrete emotions, as per the categorical model of affect. However, continuous dimensional models of affect, such as those using valence/arousal, provide significant advantages over categorical models in terms of representing complex human emotional states. In this thesis, a novel method to derive dimensional annotations from typical categorical affect datasets will be proposed. The thesis further shows that dimensional and categorical emotion spaces can be jointly learnt via label fusion in order to obtain better feature representations. Experiments on in-the-wild AffectNet as well as a collection of laboratory based FER datasets suggests that jointly learning both dimensional and categorical models of affect significantly reduce redundancy and computational load whilst improving generalization performance, outperforming state-of-the-art methods for dimensional AffectNet with 0.33 and 0.30 RMSEs."]},{"key":"dc:format.checksum.md5","label":"Dc Format Checksum Md5","values":["09c2991b0db80ae7f15e42960c99d1af","e4797cd672e22c3e1e6daf3d16ebcd8d"]},{"key":"dc:title","label":"Title","values":["DEEP LEARNING FOR FOR LARGE-SCALE FACIAL EXPRESSION RECOGNITION IN THE WILD"]}]}],"canonical_facts":{"dc:creator":["NEO YUAN RONG DEXTER"],"dc:date.issued":["2021-07-28"],"dc:description.abstract":["Human facial expression recognition (FER) is a challenging task due to ambiguous facial expressions and highly skewed, imbalanced datasets. Even with recent advances in deep learning, these obstacles still remain. Facial expressions are commonly assigned discrete emotions, as per the categorical model of affect. However, continuous dimensional models of affect, such as those using valence/arousal, provide significant advantages over categorical models in terms of representing complex human emotional states. In this thesis, a novel method to derive dimensional annotations from typical categorical affect datasets will be proposed. The thesis further shows that dimensional and categorical emotion spaces can be jointly learnt via label fusion in order to obtain better feature representations. Experiments on in-the-wild AffectNet as well as a collection of laboratory based FER datasets suggests that jointly learning both dimensional and categorical models of affect significantly reduce redundancy and computational load whilst improving generalization performance, outperforming state-of-the-art methods for dimensional AffectNet with 0.33 and 0.30 RMSEs."],"dc:format.checksum.md5":["09c2991b0db80ae7f15e42960c99d1af","e4797cd672e22c3e1e6daf3d16ebcd8d"],"dc:identifier.uri":["https://scholarbank.nus.edu.sg/bitstreams/7c474e6f-ee19-4abd-86b5-967dc1af99d1/download"],"dc:relation.isreferencedby":["https://scholarbank.nus.edu.sg/handle/10635/198469"],"dc:subject":["Emotions, Deep Learning, Facial Expression Recognition"],"dc:title":["DEEP LEARNING FOR FOR LARGE-SCALE FACIAL EXPRESSION RECOGNITION IN THE WILD"],"dc:type":["Thesis"]},"updated_at":"2026-07-24T03:31:51Z"}