The DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index.
Separating Style and Content with Bilinear Models
Abstract
Perceptual systems routinely separate “content” from “style,” classifying familiar words spoken in an unfamiliar accent, identifying a font or handwriting style across letters, or recognizing a familiar face or object seen under unfamiliar viewing conditions. Yet a general and tractable computational model of this ability to untangle the underlying factors of perceptual observations remains elusive (Hofstadter, 1985). Existing factor models (Mardia, Kent, & Bibby, 1979; Hinton & Zemel, 1994; Ghahramani, 1995; Bell & Sejnowski, 1995; Hinton, Dayan, Frey, & Neal, 1995; Dayan, Hinton, Neal, & Zemel, 1995; Hinton & Ghahramani, 1997) are either insufficiently rich to capture the complex interactions of perceptually meaningful factors such as phoneme and speaker accent or letter and font, or do not allow efficient learning algorithms. We present a general framework for learning to solve two-factor tasks using bilinear models, which provide sufficiently expressive representations of factor interactions but can nonetheless be fit to data using efficient algorithms based on the singular value decomposition and expectation-maximization. We report promising results on three different tasks in three different perceptual domains: spoken vowel classification with a benchmark multi-speaker database, extrapolation of fonts to unseen letters, and translation of faces to novel illuminants.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“GeAN also uses the bilinear model [29,30] to distil the knowledge interwoven in these representations by allowing each node and edge to collaboratively modulate each other's contributions through a shared learning space.”
From TVGeAN: Tensor Visibility Graph-Enhanced Attention Network for Versatile Multivariant Time Series Learning Tasks (2024) · cites this paper as [29]
Citations
Cites
Maximum Likelihood from Incomplete Data Via the EM Algorithm
Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.2517-6161.1977.tb01600.x · Cited by 50,444
Neural networks for pattern recognition
Bishop, C. (1994) · Choice Reviews Online · doi:10.5860/choice.31-5500 · Cited by 18,537
Press, W. H., Teukolsky, S. A., Vetterling, W. T., & Flannery, B. P. (1994) · Cited by 15,411
Turk, M., & Pentland, A. (1991) · Journal of Cognitive Neuroscience · doi:10.1162/jocn.1991.3.1.71 · Cited by 13,733
(1973) · CERN Document Server (European Organization for Nuclear Research) · Cited by 12,895
Numerical recipes in Pascal: the art of scientific computing
(1990) · Choice Reviews Online · doi:10.5860/choice.27-5158 · Cited by 12,040
An Information-Maximization Approach to Blind Separation and Blind Deconvolution
Bell, A. J., & Sejnowski, T. J. (1995) · Neural Computation · doi:10.1162/neco.1995.7.6.1129 · Cited by 8,986
Shape and motion from image streams under orthography: a factorization method
Tomasi, C., & Takeo Kanade, A. (1992) · International Journal of Computer Vision · doi:10.1007/bf00129684 · Cited by 2,879
Matrix Differential Calculus with Applications in Statistics and Econometrics.
Magnus, J. R., & Neudecker, H. (1988) · Biometrics · doi:10.2307/2531754 · Cited by 2,729
Classical Electrodynamics (2nd edn)
Brereton, M. (1976) · Physics Bulletin · doi:10.1088/0031-9112/27/8/041 · Cited by 2,612
Cited by
Sorted by citations
Items 201–300 of 839
Component-Based Recognition of Facesand Facial Expressions
Taheri, S., Patel, V. M., & Chellappa, R. (2013) · IEEE Transactions on Affective Computing · doi:10.1109/t-affc.2013.28 · Cited by 47
“Index Terms—Simultaneous face and expression recognition, face recognition, expression recognition, image separation, sparse representation, dictionary learning Ç”
BackgroundModeling the joint density of two images under a variety of transformations
Susskind, J. M., Hinton, G. E., Memisevic, R., & Pollefeys, M. (2011) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2011.5995541 · Cited by 47
“have also been addressed with bilinear models (for example, [23], [18], [4]).”
Tateno, S., Liu, H., & Ou, J. (2020) · Sensors · doi:10.3390/s20205807 · Cited by 48
“Suppose that the EMG signal is Y ∈ R C where c ∈ 1 ∼ C is the channel serial number of the EMG signal, the subject serial number is u ∈ 1 ∼ U the motion serial number is m ∈ 1 ∼ M and the data serial number for one motion is n ∈ 1 ∼ Therefore, the problem of fitting the bilinear model can be…”
BackgroundSimultaneous Unsupervised Learning of Disparate Clusterings
Jain, P., Meka, R., & Dhillon, I. S. (2008) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.10007 · Cited by 47
“discovering two-factor structures from the observed data was suggested in 20].”
BackgroundA general bilinear model to describe growth or decline time profiles
Buchwald, P. (2006) · Mathematical Biosciences · doi:10.1016/j.mbs.2006.08.013 · Cited by 48
Anytime Recognition of Objects and Scenes
Karayev, S., Fritz, M., & Darrell, T. (2014) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2014.80 · Cited by 50
“C on en t Separate from the application domain of vision, some machine learning research has attempted to separate style from content (Tenenbaum and Freeman 2000).”
BackgroundFacial Expression Analysis Using Nonlinear Decomposable Generative Models
Lee, C.-S., & Elgammal, A. (2005) · Lecture notes in computer science · doi:10.1007/11564386_3 · Cited by 45
“Bilinear models were applied to decompose person-dependent factor and the pose-dependent factor as the style and content from pose-aligned face images of different people [ 20 ] and facial expression synthesis [4].”
MethodCommon Feature Discriminant Analysis for Matching Infrared Face Images to Optical Face Images
Li, Z., Gong, D., Qiao, Y., et al. (2014) · IEEE Transactions on Image Processing · doi:10.1109/tip.2014.2315920 · Cited by 45
“[10] applied the Bilinear Model (BLM) by Singular Value Decomposition (SVD) to develop a common content (associated with identity) space for a set of different styles (corresponding to modalities).”
BackgroundMethodGuiding InfoGAN with Semi-supervision
Spurr, A., Aksan, E., & Hilliges, O. (2017) · Lecture notes in computer science · doi:10.1007/978-3-319-71249-9_8 · Cited by 49
“An example of an early approach is supervised bi-linear models [27], separating style from the content.”
MethodCombining Factorization Model and Additive Forest for Collaborative Followee Recommendation
Chen, T., Tang, L., Liu, Q., et al. (2012) · Cited by 44
Manifold Learning for Medical Image Registration, Segmentation, and Classification
Aljabar, P., Wolz, R., & Rueckert, D. (2012) · Advances in bioinformatics and biomedical engineering book series · doi:10.4018/978-1-4666-0059-1.ch017 · Cited by 43
Efficient discrete latent semantic hashing for scalable cross-modal retrieval
Lu, X. C., Zhu, L., Cheng, Z., et al. (2018) · Signal Processing · doi:10.1016/j.sigpro.2018.09.007 · Cited by 44
Prognostics of surgical site infections using dynamic health data
Ke, C., Jin, Y., Evans, H. L., et al. (2016) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2016.10.021 · Cited by 43
“A general framework was presented in [24] for solving two-factor tasks using bilinear models, which can characterize factor interactions and employ efficient algorithms based on the singular value decomposition and expectation-maximization.”
BackgroundMethodAttentional Kernel Encoding Networks for Fine-Grained Visual Categorization
Hu, Y., Yang, Y., Zhang, J., Cao, X., & Zhen, X. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.2978115 · Cited by 44
“It was first proposed by Tanenbaum and Freeman [37] to model two-factor variations.”
MethodGait style and gait content: bilinear models for gait recognition using gait re-sampling
Lee, C.-S., & Elgammal, A. (2004) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/afgr.2004.1301522 · Cited by 43
“sition can be achieved by singular value decomposition (SVD) as was shown in [ 20 ].”
BackgroundMethodInfluentialDeep Fusion: An Attention Guided Factorized Bilinear Pooling for Audio-video Emotion Recognition
Zhang, Y., Wang, Z.-R., & Du, J. (2019) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2019.8851942 · Cited by 41
“In contrast to linear models, bilinear pooling [18] has recently been used to integrate different CNN features for fine-grained image recognition [19].”
MethodNumerical linear algebra in data mining
Eldén, L. (2006) · Acta Numerica · doi:10.1017/s0962492906240017 · Cited by 42
“It is not uncommon in the data mining/machine learning literature for data compression and rank reduction problems to be presented as matrix problems, while they can in fact be considered as tensor approximation problems: for examples see Tenenbaum and Freeman (2000) and Ye (2005).”
BackgroundA comprehensive review of visual–textual sentiment analysis from social media networks
Al-Tameemi, I. K. S., Feizi‐Derakhshi, M., Pashazadeh, S., & Asadpour, M. (2024) · Journal of Computational Social Science · doi:10.1007/s42001-024-00326-y · Cited by 50
“Bilinear pooling, also called order pooling [163], is a technique for integrating visual and textual information into a single representation space by calculating the outer product between the feature vectors of these modalities, allowing all the elements in both vectors to interact in a…”
MethodCanonical Correlation Analysis With L 2,1-Norm for Multiview Data Representation
Xu, M., Zhu, Z., Zhang, X., et al. (2019) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2019.2904753 · Cited by 41
“…have been developed [6], [8], [11]–[20], of which the most popular ones include canonical correlation analysis (CCA) [11], bilinear model (BLM) [12], multioutput regularized feature projection (MORP) [15], partial least square (PLS) [16], and deep neural networks-based (DNNs-based) approaches…”
BackgroundAhmed, S., Frikha, M., Hussein, T. D. H., et al. (2021) · BioMed Research International · doi:10.1155/2021/6621540 · Cited by 40
“The authors in [52] introduced a considerable idea by bilinear generative models to decompose orthogonal factors.”
BackgroundRetractedA Structured Model of Video Reproduces Primary Visual Cortical Organisation
Berkes, P., Turner, R. E., & Sahani, M. (2009) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1000495 · Cited by 42
“Tenenbaum and Freeman [36] showed that the rich nonlinear interactions between these two factors facilitated classification and extrapolation in a series of experiments using spoken vowels, letters in different fonts, and faces in different poses.”
BackgroundMethodMultimodal Attention Network for Continuous-Time Emotion Recognition Using Video and EEG Signals
Choi, D. Y., Kim, D.-H., & Song, B. C. (2020) · IEEE Access · doi:10.1109/access.2020.3036877 · Cited by 39
“Firstly, three different fusion methods were evaluated: feature concatenation & FC layer, bilinear pooling [51], and low-rank decomposition.”
BackgroundMethodMutual Component Analysis for Heterogeneous Face Recognition
Li, Z., Gong, D., Li, Q., et al. (2016) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/2807705 · Cited by 39
Group-based Learning of Disentangled Representations with Generalizability for Novel Contents
Hosoya, H. (2019) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2019/348 · Cited by 39
Entity Slot Filling for Visual Captioning
Bin, Y., Ding, Y., Peng, B., Peng, L., Yang, Y., & Chua, T. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3063297 · Cited by 39
What is the other 85% of V1 doing?
Olshausen, B. A., & Field, D. J. (2004) · Cited by 42
Joint Latent Subspace Learning and Regression for Cross-Modal Retrieval
Wu, J., Lin, Z., & Zha, H. (2017) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3077136.3080678 · Cited by 38
“As PLS [13], BLM [16] and CCA [3] only utilize the pairwise closeness, their performance is much lower than others that use the class information.”
BackgroundMethodResultInfluentialSynergistic Instance-Level Subspace Alignment for Fine-Grained Sketch-Based Image Retrieval
Li, K., Pang, K., Song, Y.-Z., et al. (2017) · IEEE Transactions on Image Processing · doi:10.1109/tip.2017.2745106 · Cited by 38
“1] CCA [14] BLM [37] (a) GMMFA [19] Our FG−SA [Eq.”
BackgroundMethodInfluentialLearning visual and textual representations for multimodal matching and classification
Liu, Y., Liu, L., Guo, Y., & Lew, M. S. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.07.001 · Cited by 39
Facial Action Transfer with Personalized Bilinear Regression
Huang, D., & la Torre, F. D. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-33709-3_11 · Cited by 37
Bilinear Models for Spatio-Temporal Point Distribution Analysis
Hoogendoorn, C., Sukno, F. M., Ordás, S., & Frangi, A. F. (2009) · International Journal of Computer Vision · doi:10.1007/s11263-009-0212-6 · Cited by 37
“1 we already mentioned the relation between the bilinear models as they are formulated in Tenenbaum and Freeman (2000) and here, and tensor space decomposition (De Lathauwer et al. 2000).”
BackgroundMethodInfluentialGyawali, P., Horáček, B. M., Sapp, J. L., & Wang, L. (2019) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2019.2939138 · Cited by 35
“other generative factors in the data? In [11], a bilinear model was presented to capture sufficiently expressive representations of factors of variations in the data, demonstrated in separating”
MethodResonator Networks, 2: Factorization Performance and Capacity Compared to Optimization-Based Methods
Kent, S. J., Frady, E. P., Sommer, F. T., & Olshausen, B. A. (2020) · Neural Computation · doi:10.1162/neco_a_01329 · Cited by 41
“Bilinear models of style and content (Tenenbaum & Freeman, 2000) was an inspiration for us in deciding to work on factorization problems.”
Multidimensional data classification with chordal distance based kernel and Support Vector Machines
Cyganek, B., Krawczyk, B., & Woźniak, M. (2015) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2015.08.001 · Cited by 37
Park, S. W., & Savvides, M. (2007) · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · doi:10.1109/tsmcb.2007.904575 · Cited by 37
“Tenenbaum and Freeman [8] assumed the limited numbers of Gaussian mixtures to model two facial factors, people and poses, and applied the expectation-maximization algorithm to obtain the parameters of the Gaussian mixture model.”
MethodBiomedical relation extraction via knowledge-enhanced reading comprehension
Jing, C., Hu, B., Peng, W., Chen, Q., & Tang, B. (2022) · BMC Bioinformatics · doi:10.1186/s12859-021-04534-5 · Cited by 39
“A bilinear [20] operation is employed to calculate the attention weights between hidden representation h Then, each relation representation r t is aligned to each hidden state h i .”
MethodThe Role of Manifold Learning in Human Motion Analysis
Elgammal, A., & Lee, C.-S. (2007) · Computational imaging and vision · doi:10.1007/978-1-4020-6693-1_2 · Cited by 36
“The pioneering work of Tenenbaum and Freeman [76] formulated the separation of style and content using a bilinear model framework [48].”
BackgroundMethodCurve Style Analysis in a Set of Shapes
Li, H., Zhang, H., Wang, Y., Cao, J., Shamir, A., & Cohen‐Or, D. (2013) · Computer Graphics Forum · doi:10.1111/cgf.12015 · Cited by 37
Transfer learning to account for idiosyncrasy in face and body expressions
Romera‐Paredes, B., Aung, M. H., Pontil, M., Bianchi‐Berthouze, N., Williams, A. C. D. C., & Watson, P. (2013) · International Conference on Automatic Face and Gesture Recognition · doi:10.1109/fg.2013.6553779 · Cited by 36
“A different approach was proposed by [16] which decouples the contribution caused by identity and expression using bilinear models [22].”
BackgroundTwo-Point Gait: Decoupling Gait from Body Shape
Lombardi, S., Nishino, K., Makihara, Y., et al. (2013) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2013.133 · Cited by 36
“[36] J. B. Tenenbaum and W. T. Freeman.”
InfluentialShao, J., Wang, L., Zhao, Z., et al. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.06.047 · Cited by 36
Memisevic, R., & Hinton, G. E. (2004) · Cited by 36
Low-rank Bilinear Pooling for Fine-Grained ClassificationPreprint
Kong, S., & Fowlkes, C. C. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.05109 · Cited by 36
Attention-Enhanced Multimodal Learning for Conceptual Design Evaluations
Song, B., Miller, S. R., & Ahmed, F. (2023) · Journal of Mechanical Design · doi:10.1115/1.4056669 · Cited by 46
Tang, Y., Salakhutdinov, R., & Hinton, G. E. (2013) · International Conference on Machine Learning · Cited by 35
Generalized Orderless Pooling Performs Implicit Salient Matching
Simon, M., Gao, Y., Darrell, T., Denzler, J., & Rodner, E. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.531 · Cited by 35
Multimodal Deep Network Embedding With Integrated Structure and Attribute Information
Zheng, C., Pan, L., & Wu, P. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2019.2920267 · Cited by 35
“cal methods, canonical correlation analysis, partial least squares, and bilinear models, and their variants [28]–[30]”
MethodPredicting Serves in Tennis using Style Priors
Wei, X., Lucey, P., Morgan, S., et al. (2015) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/2783258.2788598 · Cited by 35
“In vision, the seminal work on separating style from content was by Tenenbaum and Freeman [28] where they used a bilinear model for a host of recognition tasks.”
MethodDeep ancient Roman Republican coin classification via feature fusion and attention
Anwar, H., Anwar, S., Zambanini, S., & Porikli, F. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2021.107871 · Cited by 38
Bilinear Active Appearance Models
González–Mora, J. L., la Torre, F. D., Murthi, R., et al. (2007) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2007.4409185 · Cited by 34
Drori, I., Cohen‐Or, D., & Yeshurun, H. (2003) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2003.1211464 · Cited by 34
Generative Adversarial Networks
Arora, J. K., Tushir, M., Kherwa, P., & Rathee, S. (2023) · Wiley eBooks · doi:10.1002/9781119879862.ch10 · Cited by 39
Approximated Bilinear Modules for Temporal Modeling
Zhu, X., Xu, C., Hui, L., Lu, C., & Tao, D. (2019) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00359 · Cited by 33
Quality Guided Sketch-to-Photo Image Synthesis
Osahor, U., Kazemi, H., Dabouei, A., et al. (2020) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw50498.2020.00418 · Cited by 33
“The content loss function is a good alternative to solely using L1 or L2 losses, as it gives better and sharper high quality reconstruction images [38].”
MethodLearning Parametric Dynamic Movement Primitives from Multiple Demonstrations
Matsubara, T., Hyon, S.-H., & Morimoto, J. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-17537-4_43 · Cited by 33
Illumination-robust face recognition using ridge regressive bilinear models
Shin, D.-S., Lee, H.-S., & Kim, D. (2007) · Pattern Recognition Letters · doi:10.1016/j.patrec.2007.08.013 · Cited by 33
Appearance factorization based facial expression recognition and synthesis
Abboud, B., & Davoine, F. (2004) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr.2004.1333729 · Cited by 33
Face recognition under large illumination variations using homomorphic filtering in spatial domain
Shahamat, H., & Pouyan, A. A. (2014) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2014.02.007 · Cited by 32
Álvarez-Tuñón, O., Jardón, A., & Balaguer, C. (2019) · Sensors · doi:10.3390/s19245497 · Cited by 32
“The first works to introduce these terms can be found in the area of language processing, concretely in the area of character recognition, separating style and content with bilinear models [31].”
BackgroundHierarchical Self-Distilled Feature Learning for Fine-Grained Visual Categorization
Hu, Y., Jiang, X., Liu, X., Luo, X., Hu, Y. G., Cao, X., Zhang, B., & Zhang, J. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3124135 · Cited by 32
“As far as we know, bilinear pooling [35], exploiting pairwise interaction between feature channels, is one of the earliest encoding approaches involved in FGVC.”
Multi-feature canonical correlation analysis for face photo-sketch image retrieval
Gong, D., Li, Z., Liu, J., et al. (2013) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/2502081.2502162 · Cited by 32
“Categories and Subject Descriptors I.4.9 [IMAGE PROCESSING AND COMPUTER VISION]: Applications; I.5.4 [PATTERN RECOGNITION]: Applications General Terms Algorithms Keywords Image retrieval, face photo-sketch matching, canonical correlation analysis.”
BackgroundInfluentialUsing Bilinear Models for View-invariant Action and Identity Recognition
Fabio Cuzzolin, A. (2006) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2006.323 · Cited by 32
Kernel-based multifactor analysis for image synthesis and recognition
Li, Y., Du, Y., & Lin, X. (2005) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2005.131 · Cited by 32
“Similar to the iterative ‘translation’ procedure of the bilinear model [1], uV, uL and uP can be updated iteratively until they converge.”
BackgroundMethodInfluentialXiong, K., Jiang, J., Pan, Y., Yang, Y., Chen, X., & Yu, Z. (2022) · Sensors · doi:10.3390/s22145322 · Cited by 34
“Furthermore, the bilinear pool captured the relationship between features by obtaining the outer product of two vectors to model the higher-order statistical information of the features [50].”
MethodSelf-Attentive Associative MemoryPreprint
Lê, H., Tran, T., & Venkatesh, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.03519 · Cited by 34
“SAM as bi-linear model Bi-linear pooling produces output from two input vectors by considering all pairwise bit interactions and thus can be implemented by means of outer product [34].”
Compositional Dictionaries for Domain Adaptive Face Recognition
Qiu, Q., & Chellappa, R. (2015) · IEEE Transactions on Image Processing · doi:10.1109/tip.2015.2479456 · Cited by 30
“In [14], a 2-mode SVD decomposition is performed.”
BackgroundMethodFace recognition using Intrinsicfaces
Wang, Y., & Wu, Y. (2010) · Pattern Recognition · doi:10.1016/j.patcog.2010.05.021 · Cited by 30
Satellite recognition and pose estimation using homeomorphic manifold analysis
Zhang, H., Jiang, Z., & Elgammal, A. (2015) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2014.130744 · Cited by 30
“, BK for K satellites, the category/style parameters can be factorized by fitting an asymmetric bilinear model [27] to the coefficients’ tensor.”
MethodLearning temporal attention in dynamic graphs with bilinear interactions
Knyazev, B. A., Augusta, C., & Taylor, G. W. (2021) · PLoS ONE · doi:10.1371/journal.pone.0247936 · Cited by 30
“As a result, bilinear layers have some useful properties, such as separability [42].”
BackgroundMulti-view face recognition based on tensor subspace analysis and view manifold modeling
Gao, X., & Tian, C. (2009) · Neurocomputing · doi:10.1016/j.neucom.2009.06.001 · Cited by 30
“Thus, a uniform multi-view face model is achieved to deal with Available online 17 June 2009 Keywords: Face recognition Tensor subspace analysis View manifold Manifold learning 12/$ - see front matter & 2009 Elsevier B.V.”
Li, X., Chen, L., Wang, L., Wu, P., & Tong, W. (2018) · IEEE Access · doi:10.1109/access.2018.2872695 · Cited by 30
“For example, bilinear models [10] separate style and content within face images, multi-view perceptron [11] separate face identity and view point.”
BackgroundTowards Scalable View-Invariant Gait Recognition: Multilinear Analysis for Gait
Lee, C.-S., & Elgammal, A. (2005) · Lecture notes in computer science · doi:10.1007/11527923_41 · Cited by 30
“Multilinear model is a generalization of linear model (one-factor models) and bilinear model (two-factor models) [ 25 ] into higher-order tensor decomposition (multi-factor models).”
MethodUser Action Interpretation for Online Content Optimization
Bian, J., Dong, A., He, X., Reddy, S., & Chang, Y. (2012) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2012.130 · Cited by 29
“where W is a jxij by jzjj matrix, which is also known as a bilinear model and has been extensively studied in the literature [9], [33]; sij represents the indicator of the correlation between the user and the item, which can be conveniently related to the response, rij (i.”
BackgroundStructural Laplacian Eigenmaps for Modeling Sets of Multivariate Sequences
Lewandowski, M., Makris, D., Velastín, S. A., & Nebel, J. (2013) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2013.2277664 · Cited by 29
Mo, Y., & Sun, S. (2014) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2014.6889523 · Cited by 29
“Other typical approaches include bilinear model (BLM) [6] and partial least squares (PLS) [7].”
MethodConsistency-aware and Inconsistency-aware Graph-based Multi-view Clustering
Horie, M., & Kasai, H. (2020) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco47968.2020.9287516 · Cited by 29
Consistent image analogies using semi-supervised learning
Cheng, L., Vishwanathan, S. V. N., & Zhang, X. (2008) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2008.4587364 · Cited by 29
VISALOGY: Answering Visual Analogy QuestionsPreprint
Sadeghi, F., Zitnick, C. L., & Farhadi, A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1510.08973 · Cited by 29
“Related to analogies is learning how to separate category and style properties in images, which has been studied using bilinear models [17].”
BackgroundFrom A to Z: Supervised Transfer of Style and Content Using Deep Neural Network GeneratorsPreprint
Upchurch, P., Snavely, N., & Bala, K. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1603.02003 · Cited by 48
“Tenenbaum and Freeman [28] propose an extrapolation method that produces letters of a font given only examples of other letters of the same font.”
MethodGuo, Y., Ge, H., & Jin-hong, L. (2023) · Frontiers in Computer Science · doi:10.3389/fcomp.2023.1159063 · Cited by 35
“This is accomplished by computing the outer product of both vectors, which makes it possible for multiplicative interactions to take place among all of the elements in both vectors (Tenenbaum and Freeman, 2000).”
BackgroundMultiview Face Recognition: From TensorFace to V-TensorFace and K-TensorFace
Tian, C., Fan, G., Gao, X., et al. (2012) · IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · doi:10.1109/tsmcb.2011.2169452 · Cited by 28
“identity-independent view representation, bilinear analysis was introduced to separate “style” (view) and “content” (identity) factors [14].”
MethodFacial Expression Recognition in the Presence of Speech Using Blind Lexical Compensation
Mariooryad, S., & Busso, C. (2015) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2015.2490070 · Cited by 28
“We use 814 of these segments.”
BackgroundMethodInfluentialFuzzy Style K-Plane Clustering
Gu, S., Nojima, Y., Ishibuchi, H., et al. (2020) · IEEE Transactions on Fuzzy Systems · doi:10.1109/tfuzz.2020.2979676 · Cited by 28
“However, in many practical scenarios, data samples in the given dataset often appear together from their respective fields or origins [17]–[20], [44]–[45], and each field/origin implicitly or explicitly shares a homogeneous style such that data samples therein become statistically depen-”
BackgroundInfluentialLearnable pooling weights for facial expression recognition
Mahmoudi, M. A., Chetouani, A., Boufera, F., et al. (2020) · Pattern Recognition Letters · doi:10.1016/j.patrec.2020.09.001 · Cited by 28
“Inspired from the second order pooling model introduced by Tenenbaum and Freeman [29] , it consists of using two CNNs as feature extractors and combine their outputs by multi-plying each location of the resulting feature maps using the outer product.”
Matsubara, T., Hyon, S., & Morimoto, J. (2011) · IEEE International Conference on Robotics and Biomimetics · doi:10.1109/robio.2011.6181317 · Cited by 27
“(3) with respect to both Z and X although the interaction matrix Wk is unknown [17].”
BackgroundMethodStyle learning and transferring for facial animation editing
Ma, X., Le, B. H., & Deng, Z. (2009) · ACM SIGGRAPH/Eurographics Symposium on Computer Animation (SCA) · doi:10.1145/1599470.1599486 · Cited by 28
“In addition, the bilinear model [ TF00 ] and the multilinear model have been successfully adapted for the transformation and transferring of facial expressions [CDB02,VBPP05].”
MethodBilinear factorisation for facial expression analysis and synthesis
Abboud, B., & Davoine, F. (2005) · IEE Proceedings - Vision Image and Signal Processing · doi:10.1049/ip-vis:20045060 · Cited by 28
“mined identity and expression is detailed in [ 10 ].”
BackgroundMethodInfluentialAn Adversarial Neuro-Tensorial Approach for Learning Disentangled Representations
Wang, M., Shu, Z., Cheng, S., Panagakis, Y., Samaras, D., & Zafeiriou, S. (2019) · International Journal of Computer Vision · doi:10.1007/s11263-019-01163-7 · Cited by 27
Huang, K., Jiang, H., & Zhang, X.-Y. (2017) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2017.2751062 · Cited by 28
Globally optimal bilinear programming for computer vision applications
Chandraker, M. K., & Kriegman, D. J. (2008) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr.2008.4587846 · Cited by 27
“Bilinearity is an oft-encountered phenomenon in computer vision, since observables in a vision system commonly arise due to an interaction between physical aspects wellapproximated by linear models [27].”
BackgroundMethodInfluentialSpeech Enhancement, Modeling and Recognition- Algorithms and Applications
Ramakrishnan, S. (2012) · InTech eBooks · doi:10.5772/2391 · Cited by 27
Automatic Generation of Typographic Font From Small Font Subset
Miyazaki, T., Tsuchiya, T., Sugaya, Y., et al. (2019) · IEEE Computer Graphics and Applications · doi:10.1109/mcg.2019.2931431 · Cited by 27
“Tenenbaum and Freeman used complete fonts as references to form characters from the generation results [19].”
BackgroundJoint Object and Pose Recognition Using Homeomorphic Manifold Analysis
Zhang, H., El-Gaaly, T., Elgammal, A., et al. (2013) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v27i1.8634 · Cited by 26
“(Tenenbaum and Freeman 2000) formulated the separation of style and content based on a bilinear model.”
BackgroundImproving human action recognition by jointly exploiting video and WiFi clues
Guo, J., Shi, M., Zhu, X., et al. (2021) · Neurocomputing · doi:10.1016/j.neucom.2020.11.074 · Cited by 26
Cross-Domain Multi-task Learning for Object Detection and Saliency Estimation
Khattar, A., Hegde, S., & Hebbalaguppe, R. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00403 · Cited by 26
“However, inspired by their application in visual question answering (VQA), we employ a kernel function based on the bilinear models [43].”
MethodBilinear Factorization via Augmented Lagrange Multipliers
Bue, A. D., Xavier, J., Agapito, L., et al. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-15561-1_21 · Cited by 26
“The interaction between two factors has also been generalised to several problems [15] in a learning context.”
BackgroundReal-time stylistic prediction for whole-body human motions
Matsubara, T., Hyon, S.-H., & Morimoto, J. (2011) · Neural Networks · doi:10.1016/j.neunet.2011.08.008 · Cited by 26
Expression Transfer between Photographs through Multilinear AAM's
Macêdo, I., Brazil, E. V., & Velho, L. C. P. R. (2006) · Proceedings - Brazilian Symposium on Computer Graphics and Image Processing/Proceedings · doi:10.1109/sibgrapi.2006.18 · Cited by 26
Max-Margin Heterogeneous Information Machine for RGB-D Action Recognition
Kong, Y., & Fu, Y. (2017) · International Journal of Computer Vision · doi:10.1007/s11263-016-0982-6 · Cited by 26
“Moreover, they used features in a vector form, in which spatiotemporal structures would not be well represented (Tenenbaum and Freeman 2000; Kobayashi 2014).”
BackgroundMethodLearning typographic style: from discrimination to synthesis
Baluja, S. (2017) · Machine Vision and Applications · doi:10.1007/s00138-017-0842-6 · Cited by 37
“In contrast to [48], we do not attempt to explicitlymodel style and content separately; rather, through training a learning”
Influential
Metadata
- Authors
- Tenenbaum, J. B., & Freeman, W. T.
- Affiliations
- Massachusetts Institute of Technology (US)
- Mitsubishi Electric Research Laboratories (United States) (US)
- Venue
- Neural Computation
- Publisher
- MIT Press
- Published
- 2000-06-01
- Volume
- 12
- Issue
- 6
- Pages
- 1247-1283
- ISSN
- 0899-7667 (ISSN-L)
- 1530-888X
- OpenAlex
- W2170653751
- Semantic Scholar
- 7e85f7d59e37972ec52cbabfef0512588d87f125
- Cited by
- 840
- References
- 61
- Citations by year
- 2026: 2 · 2025: 17 · 2024: 20 · 2023: 31 · 2022: 38 · 2021: 61 · 2020: 58 · 2019: 70 · 2018: 49 · 2017: 39 · 2016: 56 · 2015: 34 · 2014: 27 · 2013: 39 · 2012: 30
- FWCI
- 13.29
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Neural Computation (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face and Expression Recognition
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Keywords
- bilinear model, factor models, perceptual learning, speaker accent, singular value decomposition, expectation-maximization algorithm, factor interactions
- MeSH terms
- Algorithms, Face, Humans, Learning, Pattern Recognition, Visual, Speech Perception, Linear Models, Neural Networks, Computer
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
