Matches in SemOpenAlex for { <https://semopenalex.org/work/W2753407461> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W2753407461 endingPage "76" @default.
- W2753407461 startingPage "66" @default.
- W2753407461 abstract "Synthesizing realistic profile faces is promising for more efficiently training deep pose-invariant models for large-scale unconstrained face recognition, by populating samples with extreme poses and avoiding tedious annotations. However, learning from synthetic faces may not achieve the desired performance due to the discrepancy between distributions of the synthetic and real face images. To narrow this gap, we propose a Dual-Agent Generative Adversarial Network (DA-GAN) model, which can improve the realism of a face simulator's output using unlabeled real faces, while preserving the identity information during the realism refinement. The dual agents are specifically designed for distinguishing real v.s. fake and identities simultaneously. In particular, we employ an off-the-shelf 3D face model as a simulator to generate profile face images with varying poses. DA-GAN leverages a fully convolutional network as the generator to generate high-resolution images and an auto-encoder as the discriminator with the dual agents. Besides the novel architecture, we make several key modifications to the standard GAN to preserve pose and texture, preserve identity and stabilize training process: (i) a pose perception loss; (ii) an identity perception loss; (iii) an adversarial loss with a boundary equilibrium regularization term. Experimental results show that DA-GAN not only presents compelling perceptual results but also significantly outperforms state-of-the-arts on the large-scale and challenging NIST IJB-A unconstrained face recognition benchmark. In addition, the proposed DA-GAN is also promising as a new approach for solving generic transfer learning problems more effectively." @default.
- W2753407461 created "2017-09-15" @default.
- W2753407461 creator A5008135430 @default.
- W2753407461 creator A5010748093 @default.
- W2753407461 creator A5011866398 @default.
- W2753407461 creator A5025518432 @default.
- W2753407461 creator A5027716366 @default.
- W2753407461 creator A5033884558 @default.
- W2753407461 creator A5039765869 @default.
- W2753407461 creator A5064802483 @default.
- W2753407461 creator A5078587654 @default.
- W2753407461 creator A5085141403 @default.
- W2753407461 date "2017-01-01" @default.
- W2753407461 modified "2023-09-23" @default.
- W2753407461 title "Dual-Agent GANs for Photorealistic and Identity Preserving Profile Face Synthesis" @default.
- W2753407461 hasPublicationYear "2017" @default.
- W2753407461 type Work @default.
- W2753407461 sameAs 2753407461 @default.
- W2753407461 citedByCount "34" @default.
- W2753407461 countsByYear W27534074612017 @default.
- W2753407461 countsByYear W27534074612018 @default.
- W2753407461 countsByYear W27534074612019 @default.
- W2753407461 countsByYear W27534074612020 @default.
- W2753407461 countsByYear W27534074612021 @default.
- W2753407461 crossrefType "proceedings-article" @default.
- W2753407461 hasAuthorship W2753407461A5008135430 @default.
- W2753407461 hasAuthorship W2753407461A5010748093 @default.
- W2753407461 hasAuthorship W2753407461A5011866398 @default.
- W2753407461 hasAuthorship W2753407461A5025518432 @default.
- W2753407461 hasAuthorship W2753407461A5027716366 @default.
- W2753407461 hasAuthorship W2753407461A5033884558 @default.
- W2753407461 hasAuthorship W2753407461A5039765869 @default.
- W2753407461 hasAuthorship W2753407461A5064802483 @default.
- W2753407461 hasAuthorship W2753407461A5078587654 @default.
- W2753407461 hasAuthorship W2753407461A5085141403 @default.
- W2753407461 hasConcept C121332964 @default.
- W2753407461 hasConcept C144024400 @default.
- W2753407461 hasConcept C153180895 @default.
- W2753407461 hasConcept C154945302 @default.
- W2753407461 hasConcept C205711294 @default.
- W2753407461 hasConcept C24890656 @default.
- W2753407461 hasConcept C2776449333 @default.
- W2753407461 hasConcept C2778355321 @default.
- W2753407461 hasConcept C2779304628 @default.
- W2753407461 hasConcept C2779803651 @default.
- W2753407461 hasConcept C31510193 @default.
- W2753407461 hasConcept C31972630 @default.
- W2753407461 hasConcept C36289849 @default.
- W2753407461 hasConcept C41008148 @default.
- W2753407461 hasConcept C76155785 @default.
- W2753407461 hasConcept C81363708 @default.
- W2753407461 hasConcept C94915269 @default.
- W2753407461 hasConceptScore W2753407461C121332964 @default.
- W2753407461 hasConceptScore W2753407461C144024400 @default.
- W2753407461 hasConceptScore W2753407461C153180895 @default.
- W2753407461 hasConceptScore W2753407461C154945302 @default.
- W2753407461 hasConceptScore W2753407461C205711294 @default.
- W2753407461 hasConceptScore W2753407461C24890656 @default.
- W2753407461 hasConceptScore W2753407461C2776449333 @default.
- W2753407461 hasConceptScore W2753407461C2778355321 @default.
- W2753407461 hasConceptScore W2753407461C2779304628 @default.
- W2753407461 hasConceptScore W2753407461C2779803651 @default.
- W2753407461 hasConceptScore W2753407461C31510193 @default.
- W2753407461 hasConceptScore W2753407461C31972630 @default.
- W2753407461 hasConceptScore W2753407461C36289849 @default.
- W2753407461 hasConceptScore W2753407461C41008148 @default.
- W2753407461 hasConceptScore W2753407461C76155785 @default.
- W2753407461 hasConceptScore W2753407461C81363708 @default.
- W2753407461 hasConceptScore W2753407461C94915269 @default.
- W2753407461 hasLocation W27534074611 @default.
- W2753407461 hasOpenAccess W2753407461 @default.
- W2753407461 hasPrimaryLocation W27534074611 @default.
- W2753407461 hasRelatedWork W1509966554 @default.
- W2753407461 hasRelatedWork W1782590233 @default.
- W2753407461 hasRelatedWork W1834627138 @default.
- W2753407461 hasRelatedWork W1916406603 @default.
- W2753407461 hasRelatedWork W1935685005 @default.
- W2753407461 hasRelatedWork W1949778830 @default.
- W2753407461 hasRelatedWork W2099471712 @default.
- W2753407461 hasRelatedWork W2145287260 @default.
- W2753407461 hasRelatedWork W2194775991 @default.
- W2753407461 hasRelatedWork W2325939864 @default.
- W2753407461 hasRelatedWork W2737047298 @default.
- W2753407461 hasRelatedWork W2798575809 @default.
- W2753407461 hasRelatedWork W2798807671 @default.
- W2753407461 hasRelatedWork W2962793481 @default.
- W2753407461 hasRelatedWork W2963073614 @default.
- W2753407461 hasRelatedWork W2963460857 @default.
- W2753407461 hasRelatedWork W2964014798 @default.
- W2753407461 hasRelatedWork W2964337551 @default.
- W2753407461 hasRelatedWork W2969985801 @default.
- W2753407461 hasRelatedWork W3099206234 @default.
- W2753407461 hasVolume "30" @default.
- W2753407461 isParatext "false" @default.
- W2753407461 isRetracted "false" @default.
- W2753407461 magId "2753407461" @default.
- W2753407461 workType "article" @default.