Matches in SemOpenAlex for { <https://semopenalex.org/work/W2970293673> ?p ?o ?g. }
- W2970293673 abstract "Facial action unit (AU) detection in the wild is a challenging problem, due to the unconstrained variability in facial appearances and the lack of accurate AU annotations. Most existing methods either depend on impractical labor-intensive labeling by experts, or inaccurate pseudo labels. In this paper, we propose an end-to-end semi-supervised unconstrained AU detection framework, which transfers accurate AU labels from a constrained source domain to an unconstrained target domain by exploiting accurate labels of AU-related facial landmarks. Specifically, we map a source image with AU label and a target image without AU label into a latent feature domain by combining source landmark-related feature with target landmark-free feature. Due to the combination of source AU-related information and target AU-free information, the latent feature domain with the transferred source AU label can be learned by maximizing the target-domain AU detection performance. Moreover, to disentangle landmark-related and landmark-free features, we introduce a novel landmark adversarial loss which can solve the multi-player minimax game in adversarial learning. Our framework can also be naturally extended for use with target-domain pseudo AU labels. Extensive experiments show that our method soundly outperforms the baselines, upper-bounds and state-of-the-art approaches on the challenging BP4D, GFT and EmotioNet benchmarks. The code for our method is available at this https URL" @default.
- W2970293673 created "2019-09-05" @default.
- W2970293673 creator A5017901486 @default.
- W2970293673 creator A5019288783 @default.
- W2970293673 creator A5065578418 @default.
- W2970293673 creator A5081933126 @default.
- W2970293673 creator A5084218062 @default.
- W2970293673 date "2019-03-25" @default.
- W2970293673 modified "2023-09-27" @default.
- W2970293673 title "Semi-Supervised Unconstrained Action Unit Detection via Latent Feature Domain" @default.
- W2970293673 cites W1509966554 @default.
- W2970293673 cites W1588539311 @default.
- W2970293673 cites W1595126664 @default.
- W2970293673 cites W1677182931 @default.
- W2970293673 cites W1795776638 @default.
- W2970293673 cites W2045472600 @default.
- W2970293673 cites W2051297709 @default.
- W2970293673 cites W2052677124 @default.
- W2970293673 cites W2099471712 @default.
- W2970293673 cites W2101545465 @default.
- W2970293673 cites W2103943262 @default.
- W2970293673 cites W2157285372 @default.
- W2970293673 cites W2421475762 @default.
- W2970293673 cites W2436394355 @default.
- W2970293673 cites W2474338411 @default.
- W2970293673 cites W2502312327 @default.
- W2970293673 cites W2519753233 @default.
- W2970293673 cites W2592922859 @default.
- W2970293673 cites W2593414223 @default.
- W2970293673 cites W2609211631 @default.
- W2970293673 cites W2734289922 @default.
- W2970293673 cites W2737047298 @default.
- W2970293673 cites W2737559518 @default.
- W2970293673 cites W2780309588 @default.
- W2970293673 cites W2784208377 @default.
- W2970293673 cites W2792605498 @default.
- W2970293673 cites W2798484817 @default.
- W2970293673 cites W2798545058 @default.
- W2970293673 cites W2806833697 @default.
- W2970293673 cites W2807725536 @default.
- W2970293673 cites W2884460600 @default.
- W2970293673 cites W2885517064 @default.
- W2970293673 cites W2896868133 @default.
- W2970293673 cites W2905511418 @default.
- W2970293673 cites W2948303854 @default.
- W2970293673 cites W2949117887 @default.
- W2970293673 cites W2949987290 @default.
- W2970293673 cites W2950594409 @default.
- W2970293673 cites W2953226914 @default.
- W2970293673 cites W2962793481 @default.
- W2970293673 cites W2962971105 @default.
- W2970293673 cites W2963562727 @default.
- W2970293673 cites W2963789946 @default.
- W2970293673 cites W2963839617 @default.
- W2970293673 cites W2964014798 @default.
- W2970293673 cites W2964121744 @default.
- W2970293673 cites W2969059826 @default.
- W2970293673 cites W2970971581 @default.
- W2970293673 cites W2971040546 @default.
- W2970293673 cites W2981595954 @default.
- W2970293673 cites W3035906484 @default.
- W2970293673 cites W3102025332 @default.
- W2970293673 cites W3159890710 @default.
- W2970293673 cites W2995034616 @default.
- W2970293673 hasPublicationYear "2019" @default.
- W2970293673 type Work @default.
- W2970293673 sameAs 2970293673 @default.
- W2970293673 citedByCount "0" @default.
- W2970293673 crossrefType "posted-content" @default.
- W2970293673 hasAuthorship W2970293673A5017901486 @default.
- W2970293673 hasAuthorship W2970293673A5019288783 @default.
- W2970293673 hasAuthorship W2970293673A5065578418 @default.
- W2970293673 hasAuthorship W2970293673A5081933126 @default.
- W2970293673 hasAuthorship W2970293673A5084218062 @default.
- W2970293673 hasConcept C111919701 @default.
- W2970293673 hasConcept C115961682 @default.
- W2970293673 hasConcept C121332964 @default.
- W2970293673 hasConcept C126255220 @default.
- W2970293673 hasConcept C134306372 @default.
- W2970293673 hasConcept C138885662 @default.
- W2970293673 hasConcept C149728462 @default.
- W2970293673 hasConcept C153180895 @default.
- W2970293673 hasConcept C154945302 @default.
- W2970293673 hasConcept C177264268 @default.
- W2970293673 hasConcept C199360897 @default.
- W2970293673 hasConcept C2776401178 @default.
- W2970293673 hasConcept C2776760102 @default.
- W2970293673 hasConcept C2780297707 @default.
- W2970293673 hasConcept C2780791683 @default.
- W2970293673 hasConcept C33923547 @default.
- W2970293673 hasConcept C36503486 @default.
- W2970293673 hasConcept C41008148 @default.
- W2970293673 hasConcept C41895202 @default.
- W2970293673 hasConcept C43126263 @default.
- W2970293673 hasConcept C62520636 @default.
- W2970293673 hasConceptScore W2970293673C111919701 @default.
- W2970293673 hasConceptScore W2970293673C115961682 @default.
- W2970293673 hasConceptScore W2970293673C121332964 @default.
- W2970293673 hasConceptScore W2970293673C126255220 @default.
- W2970293673 hasConceptScore W2970293673C134306372 @default.