Matches in SemOpenAlex for { <https://semopenalex.org/work/W4220688397> ?p ?o ?g. }
- W4220688397 endingPage "3959" @default.
- W4220688397 startingPage "3947" @default.
- W4220688397 abstract "Recognizing facial expressions rely on facial parts’ movement (action units) such as eyes, mouth, and nose. Existing methods utilize complex subnetworks to learn part-based facial features or train neural networks with an extensively perturbed dataset. Different from existing methods, we propose a trainable end-to-end convolutional neural network for facial expression recognition. First, we propose a Local Prediction Penalty to stimulate facial expression recognition research with no part-based learning. It is a technique to punish the feature extractor’s local predictive power to coerce it to learn coarse-grained features (general facial expression). The Local Prediction Penalty forces the network to disregard predictive local signals learned from local receptive fields and instead depend on the global facial region. Second, we propose a Spatial Self-Attention method for fine-grained feature representation to learn distinct face features from pixel positions. The Spatial Self-Attention accumulates attention features at privileged positions without changing the spatial feature dimension. Lastly, we leverage a classifier to carefully combine all learned features (coarse-grained and fine-grained) for better feature representation. Extensive experiments demonstrate that our proposed methods significantly improve facial expression recognition performance." @default.
- W4220688397 created "2022-04-03" @default.
- W4220688397 creator A5030009218 @default.
- W4220688397 creator A5045032131 @default.
- W4220688397 creator A5057488763 @default.
- W4220688397 creator A5058997400 @default.
- W4220688397 creator A5080286262 @default.
- W4220688397 date "2022-08-10" @default.
- W4220688397 modified "2023-09-26" @default.
- W4220688397 title "Facial expression recognition via coarse-grained and fine-grained feature representation" @default.
- W4220688397 cites W1530420445 @default.
- W4220688397 cites W2060488580 @default.
- W4220688397 cites W2065379720 @default.
- W4220688397 cites W2198512331 @default.
- W4220688397 cites W2217426128 @default.
- W4220688397 cites W2248071184 @default.
- W4220688397 cites W2348624827 @default.
- W4220688397 cites W2470957930 @default.
- W4220688397 cites W2486769752 @default.
- W4220688397 cites W2487852963 @default.
- W4220688397 cites W2506506742 @default.
- W4220688397 cites W2600389231 @default.
- W4220688397 cites W2621864722 @default.
- W4220688397 cites W2730601341 @default.
- W4220688397 cites W2737398044 @default.
- W4220688397 cites W2793980118 @default.
- W4220688397 cites W2798553619 @default.
- W4220688397 cites W2904483377 @default.
- W4220688397 cites W2905260102 @default.
- W4220688397 cites W2915815650 @default.
- W4220688397 cites W2917996064 @default.
- W4220688397 cites W2944523338 @default.
- W4220688397 cites W2954709787 @default.
- W4220688397 cites W2960608680 @default.
- W4220688397 cites W2963092169 @default.
- W4220688397 cites W2963495494 @default.
- W4220688397 cites W2964319207 @default.
- W4220688397 cites W2981307649 @default.
- W4220688397 cites W3001196836 @default.
- W4220688397 cites W3035565904 @default.
- W4220688397 cites W3037439697 @default.
- W4220688397 cites W3097096317 @default.
- W4220688397 cites W3101998545 @default.
- W4220688397 cites W3213447082 @default.
- W4220688397 cites W4205088571 @default.
- W4220688397 doi "https://doi.org/10.3233/jifs-212022" @default.
- W4220688397 hasPublicationYear "2022" @default.
- W4220688397 type Work @default.
- W4220688397 citedByCount "0" @default.
- W4220688397 crossrefType "journal-article" @default.
- W4220688397 hasAuthorship W4220688397A5030009218 @default.
- W4220688397 hasAuthorship W4220688397A5045032131 @default.
- W4220688397 hasAuthorship W4220688397A5057488763 @default.
- W4220688397 hasAuthorship W4220688397A5058997400 @default.
- W4220688397 hasAuthorship W4220688397A5080286262 @default.
- W4220688397 hasConcept C115961682 @default.
- W4220688397 hasConcept C138885662 @default.
- W4220688397 hasConcept C153083717 @default.
- W4220688397 hasConcept C153180895 @default.
- W4220688397 hasConcept C154945302 @default.
- W4220688397 hasConcept C195704467 @default.
- W4220688397 hasConcept C2776401178 @default.
- W4220688397 hasConcept C41008148 @default.
- W4220688397 hasConcept C41895202 @default.
- W4220688397 hasConcept C53533937 @default.
- W4220688397 hasConcept C59404180 @default.
- W4220688397 hasConcept C81363708 @default.
- W4220688397 hasConcept C87335442 @default.
- W4220688397 hasConcept C95623464 @default.
- W4220688397 hasConceptScore W4220688397C115961682 @default.
- W4220688397 hasConceptScore W4220688397C138885662 @default.
- W4220688397 hasConceptScore W4220688397C153083717 @default.
- W4220688397 hasConceptScore W4220688397C153180895 @default.
- W4220688397 hasConceptScore W4220688397C154945302 @default.
- W4220688397 hasConceptScore W4220688397C195704467 @default.
- W4220688397 hasConceptScore W4220688397C2776401178 @default.
- W4220688397 hasConceptScore W4220688397C41008148 @default.
- W4220688397 hasConceptScore W4220688397C41895202 @default.
- W4220688397 hasConceptScore W4220688397C53533937 @default.
- W4220688397 hasConceptScore W4220688397C59404180 @default.
- W4220688397 hasConceptScore W4220688397C81363708 @default.
- W4220688397 hasConceptScore W4220688397C87335442 @default.
- W4220688397 hasConceptScore W4220688397C95623464 @default.
- W4220688397 hasIssue "4" @default.
- W4220688397 hasLocation W42206883971 @default.
- W4220688397 hasOpenAccess W4220688397 @default.
- W4220688397 hasPrimaryLocation W42206883971 @default.
- W4220688397 hasRelatedWork W1982770690 @default.
- W4220688397 hasRelatedWork W2275058042 @default.
- W4220688397 hasRelatedWork W2461933888 @default.
- W4220688397 hasRelatedWork W2613736958 @default.
- W4220688397 hasRelatedWork W2760085659 @default.
- W4220688397 hasRelatedWork W3093612317 @default.
- W4220688397 hasRelatedWork W3194078543 @default.
- W4220688397 hasRelatedWork W4214561993 @default.
- W4220688397 hasRelatedWork W4296131109 @default.
- W4220688397 hasRelatedWork W564581980 @default.
- W4220688397 hasVolume "43" @default.