Matches in SemOpenAlex for { <https://semopenalex.org/work/W2795486237> ?p ?o ?g. }
- W2795486237 abstract "Face multi-attribute prediction benefits substantially from multi-task learning (MTL), which learns multiple face attributes simultaneously to achieve shared or mutually related representations of different attributes. The most widely used MTL convolutional neural network is heuristically or empirically designed by sharing all of the convolutional layers and splitting at the fully connected layers for task-specific losses. However, it is improper to view all low and mid-level features for different attributes as being the same, especially when these attributes are only loosely related. In this paper, we propose a novel multi-attribute tensor correlation neural network (MTCN) for face attribute prediction. The structure shares the information in low-level features (e.g., the first two convolutional layers) but splits that in high-level features (e.g., from the third convolutional layer to the fully connected layer). At the same time, during high-level feature extraction, each subnetwork (e.g., Age-Net, Gender-Net, ..., and Smile-Net) excavates closely related features from other networks to enhance its features. Then, we project the features of the C9 layers of the fine-tuned subnetworks into a highly correlated space by using a novel tensor correlation analysis algorithm (NTCCA). The final face attribute prediction is made based on the correlation matrix. Experimental results on benchmarks with multiple face attributes (CelebA and LFWA) show that the proposed approach has superior performance compared to state-of-the-art methods." @default.
- W2795486237 created "2018-04-13" @default.
- W2795486237 creator A5047455588 @default.
- W2795486237 creator A5078793726 @default.
- W2795486237 creator A5090246160 @default.
- W2795486237 date "2018-04-09" @default.
- W2795486237 modified "2023-09-24" @default.
- W2795486237 title "A Novel Multi-Task Tensor Correlation Neural Network for Facial Attribute Prediction" @default.
- W2795486237 cites W1567822572 @default.
- W2795486237 cites W1672347394 @default.
- W2795486237 cites W1782590233 @default.
- W2795486237 cites W1834627138 @default.
- W2795486237 cites W2018282388 @default.
- W2795486237 cites W2063386797 @default.
- W2795486237 cites W2066440363 @default.
- W2795486237 cites W2073749068 @default.
- W2795486237 cites W2087914506 @default.
- W2795486237 cites W2088009615 @default.
- W2795486237 cites W2100235303 @default.
- W2795486237 cites W2105026179 @default.
- W2795486237 cites W2214352687 @default.
- W2795486237 cites W2271840356 @default.
- W2795486237 cites W2339391301 @default.
- W2795486237 cites W2491457278 @default.
- W2795486237 cites W2510725918 @default.
- W2795486237 cites W2605124728 @default.
- W2795486237 cites W2754125855 @default.
- W2795486237 cites W2765131766 @default.
- W2795486237 cites W2949483767 @default.
- W2795486237 cites W2962786991 @default.
- W2795486237 cites W2963377935 @default.
- W2795486237 cites W3103850820 @default.
- W2795486237 doi "https://doi.org/10.48550/arxiv.1804.02810" @default.
- W2795486237 hasPublicationYear "2018" @default.
- W2795486237 type Work @default.
- W2795486237 sameAs 2795486237 @default.
- W2795486237 citedByCount "0" @default.
- W2795486237 crossrefType "posted-content" @default.
- W2795486237 hasAuthorship W2795486237A5047455588 @default.
- W2795486237 hasAuthorship W2795486237A5078793726 @default.
- W2795486237 hasAuthorship W2795486237A5090246160 @default.
- W2795486237 hasBestOaLocation W27954862371 @default.
- W2795486237 hasConcept C117220453 @default.
- W2795486237 hasConcept C119857082 @default.
- W2795486237 hasConcept C124101348 @default.
- W2795486237 hasConcept C138885662 @default.
- W2795486237 hasConcept C14166107 @default.
- W2795486237 hasConcept C144024400 @default.
- W2795486237 hasConcept C153180895 @default.
- W2795486237 hasConcept C154945302 @default.
- W2795486237 hasConcept C155281189 @default.
- W2795486237 hasConcept C162324750 @default.
- W2795486237 hasConcept C178790620 @default.
- W2795486237 hasConcept C185592680 @default.
- W2795486237 hasConcept C187736073 @default.
- W2795486237 hasConcept C202444582 @default.
- W2795486237 hasConcept C2524010 @default.
- W2795486237 hasConcept C2776401178 @default.
- W2795486237 hasConcept C2779227376 @default.
- W2795486237 hasConcept C2779304628 @default.
- W2795486237 hasConcept C2780186347 @default.
- W2795486237 hasConcept C2780451532 @default.
- W2795486237 hasConcept C33923547 @default.
- W2795486237 hasConcept C36289849 @default.
- W2795486237 hasConcept C38652104 @default.
- W2795486237 hasConcept C41008148 @default.
- W2795486237 hasConcept C41895202 @default.
- W2795486237 hasConcept C81363708 @default.
- W2795486237 hasConceptScore W2795486237C117220453 @default.
- W2795486237 hasConceptScore W2795486237C119857082 @default.
- W2795486237 hasConceptScore W2795486237C124101348 @default.
- W2795486237 hasConceptScore W2795486237C138885662 @default.
- W2795486237 hasConceptScore W2795486237C14166107 @default.
- W2795486237 hasConceptScore W2795486237C144024400 @default.
- W2795486237 hasConceptScore W2795486237C153180895 @default.
- W2795486237 hasConceptScore W2795486237C154945302 @default.
- W2795486237 hasConceptScore W2795486237C155281189 @default.
- W2795486237 hasConceptScore W2795486237C162324750 @default.
- W2795486237 hasConceptScore W2795486237C178790620 @default.
- W2795486237 hasConceptScore W2795486237C185592680 @default.
- W2795486237 hasConceptScore W2795486237C187736073 @default.
- W2795486237 hasConceptScore W2795486237C202444582 @default.
- W2795486237 hasConceptScore W2795486237C2524010 @default.
- W2795486237 hasConceptScore W2795486237C2776401178 @default.
- W2795486237 hasConceptScore W2795486237C2779227376 @default.
- W2795486237 hasConceptScore W2795486237C2779304628 @default.
- W2795486237 hasConceptScore W2795486237C2780186347 @default.
- W2795486237 hasConceptScore W2795486237C2780451532 @default.
- W2795486237 hasConceptScore W2795486237C33923547 @default.
- W2795486237 hasConceptScore W2795486237C36289849 @default.
- W2795486237 hasConceptScore W2795486237C38652104 @default.
- W2795486237 hasConceptScore W2795486237C41008148 @default.
- W2795486237 hasConceptScore W2795486237C41895202 @default.
- W2795486237 hasConceptScore W2795486237C81363708 @default.
- W2795486237 hasLocation W27954862371 @default.
- W2795486237 hasOpenAccess W2795486237 @default.
- W2795486237 hasPrimaryLocation W27954862371 @default.
- W2795486237 hasRelatedWork W2175746458 @default.
- W2795486237 hasRelatedWork W2295830131 @default.
- W2795486237 hasRelatedWork W2521062615 @default.