Abstract:In this article, an effective approach is proposed to recognise the 2D+3D facial expression automatically based on orthogonal Tucker decomposition using factor priors (OTDFPFER). As a powerful technique, Tucker decomposition on the basis of the low rank approximation is often used to extract the useful information from the constructed 4D tensor composed of 3D face scans and 2D images aiming to maintain correlations and their structural information. Finding a set of projected factor matrices is our ultimate goa… Show more
“…The performance comparisons with the state-of-the-art methods (i.e., [9,16,17,19,20,23]) are shown in Table 5 on Bosphorus database by using Setup IV. From this table, we can observe that our proposed approach obtains the highest recognition accuracy, while the method [19] gains the lowest one.…”
Section: Comparison With Other Methodsmentioning
confidence: 99%
“…Among the three protocols, two expressions of happiness and surprise are achieved better recognition results because of their higher facial deformation, whereas fear expression that can be confused with other five expressions is obtained worse results and is to a great degree confused with happiness expression. Meanwhile, it can be found that sadness expression in Setup I is achieved the best recognition result among the three protocols, and even indicates a certain improvement compared with those in [16,17,20,22,25,46].…”
Section: Performance Evaluation On Bu-3dfe Databasementioning
confidence: 97%
“…and the last term from the SVD. According to the closed-form solution in (16), we can get the computation cost for updating all G (i) 's in each iteration of the order…”
“…Five state-of-the-art algorithms based on Tucker decomposition are compared with our proposed approach, which includes FERLrTC [17], WTucker [12], MR NTD [26], APG NTD [43] and OTDF-PFER [16].…”
Section: Comparison With Tucker Decomposition-based Algorithmsmentioning
confidence: 99%
“…To alleviate this issue, a more natural way to describe 3D facial expression data is using tensors, which not only maintains the spacial structure but also admits sparse representation when appropriate tensor decomposition is chosen, by employing tools from tensor analysis. At present, the existing methods using tensors to describe 3D facial expression data are mostly based on tensor decomposition [13][14][15][16][17]20,21], which have opened up a new technology direction and made some progress. However, the local structure (geometric information) of 3D tensor samples in these methods are not maintained in the low-dimensional tensors space during the dimensionality reduction.…”
“…The performance comparisons with the state-of-the-art methods (i.e., [9,16,17,19,20,23]) are shown in Table 5 on Bosphorus database by using Setup IV. From this table, we can observe that our proposed approach obtains the highest recognition accuracy, while the method [19] gains the lowest one.…”
Section: Comparison With Other Methodsmentioning
confidence: 99%
“…Among the three protocols, two expressions of happiness and surprise are achieved better recognition results because of their higher facial deformation, whereas fear expression that can be confused with other five expressions is obtained worse results and is to a great degree confused with happiness expression. Meanwhile, it can be found that sadness expression in Setup I is achieved the best recognition result among the three protocols, and even indicates a certain improvement compared with those in [16,17,20,22,25,46].…”
Section: Performance Evaluation On Bu-3dfe Databasementioning
confidence: 97%
“…and the last term from the SVD. According to the closed-form solution in (16), we can get the computation cost for updating all G (i) 's in each iteration of the order…”
“…Five state-of-the-art algorithms based on Tucker decomposition are compared with our proposed approach, which includes FERLrTC [17], WTucker [12], MR NTD [26], APG NTD [43] and OTDF-PFER [16].…”
Section: Comparison With Tucker Decomposition-based Algorithmsmentioning
confidence: 99%
“…To alleviate this issue, a more natural way to describe 3D facial expression data is using tensors, which not only maintains the spacial structure but also admits sparse representation when appropriate tensor decomposition is chosen, by employing tools from tensor analysis. At present, the existing methods using tensors to describe 3D facial expression data are mostly based on tensor decomposition [13][14][15][16][17]20,21], which have opened up a new technology direction and made some progress. However, the local structure (geometric information) of 3D tensor samples in these methods are not maintained in the low-dimensional tensors space during the dimensionality reduction.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.