Matches in SemOpenAlex for { <https://semopenalex.org/work/W2973161808> ?p ?o ?g. }
- W2973161808 endingPage "115636" @default.
- W2973161808 startingPage "115636" @default.
- W2973161808 abstract "Increasing inter-class variance and shrinking intra-class distance are two main concerns and efforts in face recognition. In this paper, we propose a new loss function termed inter-class angular margin (IAM) loss aiming to enlarge the inter-class variance. Instead of restricting the inter-class margin to be a constant in existing methods, our IAM loss adaptively penalizes smaller inter-class angles more heavily and successfully makes the angular margin between classes larger, which can significantly enhance the discrimination of facial features. The IAM loss can be readily introduced as a regularization term for the widely-used Softmax loss and its recent variants to further improve their performances. We also analyze and verify the appropriate range of the regularization hyper-parameter from the perspective of backpropagation. For illustrative purposes, our model is trained on CASIA-WebFace and tested on the LFW, CFP, YTF and MegaFace datasets; the experimental results show that the IAM loss is quite effective to improve state-of-the-art algorithms." @default.
- W2973161808 created "2019-09-19" @default.
- W2973161808 creator A5009239895 @default.
- W2973161808 creator A5026184280 @default.
- W2973161808 creator A5046860808 @default.
- W2973161808 creator A5078051038 @default.
- W2973161808 creator A5079361172 @default.
- W2973161808 date "2020-02-01" @default.
- W2973161808 modified "2023-10-16" @default.
- W2973161808 title "Inter-class angular margin loss for face recognition" @default.
- W2973161808 cites W1970748944 @default.
- W2973161808 cites W1998808035 @default.
- W2973161808 cites W2121647436 @default.
- W2973161808 cites W2167144347 @default.
- W2973161808 cites W2183341477 @default.
- W2973161808 cites W2560205181 @default.
- W2973161808 cites W2560473673 @default.
- W2973161808 cites W2737608545 @default.
- W2973161808 cites W2752632790 @default.
- W2973161808 cites W2784294927 @default.
- W2973161808 cites W2905227840 @default.
- W2973161808 cites W2924116116 @default.
- W2973161808 cites W2962922273 @default.
- W2973161808 cites W2962952900 @default.
- W2973161808 cites W2963466847 @default.
- W2973161808 cites W2963671154 @default.
- W2973161808 cites W3099206234 @default.
- W2973161808 cites W3101998545 @default.
- W2973161808 doi "https://doi.org/10.1016/j.image.2019.115636" @default.
- W2973161808 hasPublicationYear "2020" @default.
- W2973161808 type Work @default.
- W2973161808 sameAs 2973161808 @default.
- W2973161808 citedByCount "6" @default.
- W2973161808 countsByYear W29731618082020 @default.
- W2973161808 countsByYear W29731618082022 @default.
- W2973161808 countsByYear W29731618082023 @default.
- W2973161808 crossrefType "journal-article" @default.
- W2973161808 hasAuthorship W2973161808A5009239895 @default.
- W2973161808 hasAuthorship W2973161808A5026184280 @default.
- W2973161808 hasAuthorship W2973161808A5046860808 @default.
- W2973161808 hasAuthorship W2973161808A5078051038 @default.
- W2973161808 hasAuthorship W2973161808A5079361172 @default.
- W2973161808 hasBestOaLocation W29731618082 @default.
- W2973161808 hasConcept C104317684 @default.
- W2973161808 hasConcept C119857082 @default.
- W2973161808 hasConcept C121955636 @default.
- W2973161808 hasConcept C144133560 @default.
- W2973161808 hasConcept C153180895 @default.
- W2973161808 hasConcept C154945302 @default.
- W2973161808 hasConcept C185592680 @default.
- W2973161808 hasConcept C188441871 @default.
- W2973161808 hasConcept C196083921 @default.
- W2973161808 hasConcept C2776135515 @default.
- W2973161808 hasConcept C2777212361 @default.
- W2973161808 hasConcept C31510193 @default.
- W2973161808 hasConcept C33923547 @default.
- W2973161808 hasConcept C41008148 @default.
- W2973161808 hasConcept C50644808 @default.
- W2973161808 hasConcept C55493867 @default.
- W2973161808 hasConcept C63479239 @default.
- W2973161808 hasConcept C774472 @default.
- W2973161808 hasConceptScore W2973161808C104317684 @default.
- W2973161808 hasConceptScore W2973161808C119857082 @default.
- W2973161808 hasConceptScore W2973161808C121955636 @default.
- W2973161808 hasConceptScore W2973161808C144133560 @default.
- W2973161808 hasConceptScore W2973161808C153180895 @default.
- W2973161808 hasConceptScore W2973161808C154945302 @default.
- W2973161808 hasConceptScore W2973161808C185592680 @default.
- W2973161808 hasConceptScore W2973161808C188441871 @default.
- W2973161808 hasConceptScore W2973161808C196083921 @default.
- W2973161808 hasConceptScore W2973161808C2776135515 @default.
- W2973161808 hasConceptScore W2973161808C2777212361 @default.
- W2973161808 hasConceptScore W2973161808C31510193 @default.
- W2973161808 hasConceptScore W2973161808C33923547 @default.
- W2973161808 hasConceptScore W2973161808C41008148 @default.
- W2973161808 hasConceptScore W2973161808C50644808 @default.
- W2973161808 hasConceptScore W2973161808C55493867 @default.
- W2973161808 hasConceptScore W2973161808C63479239 @default.
- W2973161808 hasConceptScore W2973161808C774472 @default.
- W2973161808 hasFunder F4320321001 @default.
- W2973161808 hasFunder F4320335972 @default.
- W2973161808 hasLocation W29731618081 @default.
- W2973161808 hasLocation W29731618082 @default.
- W2973161808 hasLocation W29731618083 @default.
- W2973161808 hasOpenAccess W2973161808 @default.
- W2973161808 hasPrimaryLocation W29731618081 @default.
- W2973161808 hasRelatedWork W2001423728 @default.
- W2973161808 hasRelatedWork W2134472250 @default.
- W2973161808 hasRelatedWork W2743258233 @default.
- W2973161808 hasRelatedWork W2758063741 @default.
- W2973161808 hasRelatedWork W2807311372 @default.
- W2973161808 hasRelatedWork W2921182884 @default.
- W2973161808 hasRelatedWork W2977314777 @default.
- W2973161808 hasRelatedWork W3208883981 @default.
- W2973161808 hasRelatedWork W4307834408 @default.
- W2973161808 hasRelatedWork W4320925816 @default.
- W2973161808 hasVolume "80" @default.
- W2973161808 isParatext "false" @default.