(1) Facial expression recognition based on Local Binary Patterns: A comprehensive study

admin管理员组

文章数量:1130349

 

Keywords: Facial expression recognition, Local Binary Patterns, Support vector machine, Adaboost, Linear discriminant analysis, Linear programming.

 

Abstract:

They empirically evaluate Facial expression recognition based on statistical local features, Local Binary Patterns. It is the Support Vector Machine that turned to be the best classifier when formulating Boosted-LBP to extract the most discriminant LBP features. They also evaluate different machine learning method on different database. It is worth to mention that LBP features are stable when comes to low-resolution Facial expression recognition and compressed low-resolution video sequences captured in real-world environments.

 

Introduction:

Facial expression recognition remains difficult since it is subtle, variable and complex. 

Geometric represent the shape and locations of facial components, which are extracted to form a feature vector that represent the face geometry. (Valstar et al. [1,2]have proved geometric-based method performs better than appearance-based approaches in AUs recognition) .

However, Geometric-based method requires accurate and reliable facial feature detection and tracking.

Appearance-based methods, image filters, such as Gabor wavelets, is applied not only whole face,but also specific face-regions to extract the appearance changes [3, 4, 5, 6, 7]. Nevertheless, it is both time and memory intensive to convolve face images to extract multi-scale and multi-orientational coefficients.

In this paper, they empirically study Facial expression recognition based on LBP, which is tolerate against illumination changes and simple at computation.They exploited different machine learning methods based LBP features, furthermore,they formulate Boost-LBP by learning the most discriminative LBP features with AdaBoost.

 

Main contributions :

Study LBP for Facial expression recognition with different classifier on large database.

Investigate LBP on low-resolution Facial expression recognition.

Formulate Boost-LBP by learning the most discriminative LBP histogram.

 

References

[1] M. Valstar, I. Patras, M. Pantic, Facial action unit detection using probabilistic actively learned support vector machines on tracked facial point data, in: IEEE Conference on Computer Vision and Pattern Recognition Workshop, vol. 3, 2005, pp. 76–84.

[2] M. Valstar, M. Pantic, Fully automatic facial action unit detection and temporal analysis, in: IEEE Conference on Computer Vision and Pattern Recognition Workshop, 2006, p. 149.

[3] M.J. Lyons, J. Budynek, S. Akamatsu, Automatic classification of single facial images, IEEE Transactions on Pattern Analysis and Machine Intelligence 21 (12) (1999) 1357–1362.

[4] G. Donato, M. Bartlett, J. Hager, P. Ekman, T.Sejnowski, Classifying facial actions, IEEE Transactions on Pattern Analysis and Machine Intelligence 21 (10) (1999) 974–989.

[5] M.S. Bartlett, G. Littlewort, M. Frank, C. Lainscsek,I. Fasel, J. Movellan, Recognizing facial expression: machine learning and application to spontaneous behavior, in: IEEE Conference on Computer Vision andPattern Recognition (CVPR), 2005.

[6] Z. Zhang, M.J. Lyons, M. Schuster, S. Akamatsu,Comparison between geometry-based and Gabor-wavelets-based facial expression recognition using multi-layer perceptron, in: IEEE International Conference on Automatic Face & Gesture Recognition (FG), 1998.

[7] Y. Tian, Evaluation of face resolution for expression analysis, in: CVPR Workshop on Face Processing in Video, 2004.

[8] C. Shan, S. Gong, and P.W. McOwan, “Facial Expression Recognition based on Local Binary Patterns: A Comprehensive Study,” Image Vision Computing., vol. 27, no. 6, May 2009, pp. 803-816.

 

Keywords: Facial expression recognition, Local Binary Patterns, Support vector machine, Adaboost, Linear discriminant analysis, Linear programming.

 

Abstract:

They empirically evaluate Facial expression recognition based on statistical local features, Local Binary Patterns. It is the Support Vector Machine that turned to be the best classifier when formulating Boosted-LBP to extract the most discriminant LBP features. They also evaluate different machine learning method on different database. It is worth to mention that LBP features are stable when comes to low-resolution Facial expression recognition and compressed low-resolution video sequences captured in real-world environments.

 

Introduction:

Facial expression recognition remains difficult since it is subtle, variable and complex. 

Geometric represent the shape and locations of facial components, which are extracted to form a feature vector that represent the face geometry. (Valstar et al. [1,2]have proved geometric-based method performs better than appearance-based approaches in AUs recognition) .

However, Geometric-based method requires accurate and reliable facial feature detection and tracking.

Appearance-based methods, image filters, such as Gabor wavelets, is applied not only whole face,but also specific face-regions to extract the appearance changes [3, 4, 5, 6, 7]. Nevertheless, it is both time and memory intensive to convolve face images to extract multi-scale and multi-orientational coefficients.

In this paper, they empirically study Facial expression recognition based on LBP, which is tolerate against illumination changes and simple at computation.They exploited different machine learning methods based LBP features, furthermore,they formulate Boost-LBP by learning the most discriminative LBP features with AdaBoost.

 

Main contributions :

Study LBP for Facial expression recognition with different classifier on large database.

Investigate LBP on low-resolution Facial expression recognition.

Formulate Boost-LBP by learning the most discriminative LBP histogram.

 

References

[1] M. Valstar, I. Patras, M. Pantic, Facial action unit detection using probabilistic actively learned support vector machines on tracked facial point data, in: IEEE Conference on Computer Vision and Pattern Recognition Workshop, vol. 3, 2005, pp. 76–84.

[2] M. Valstar, M. Pantic, Fully automatic facial action unit detection and temporal analysis, in: IEEE Conference on Computer Vision and Pattern Recognition Workshop, 2006, p. 149.

[3] M.J. Lyons, J. Budynek, S. Akamatsu, Automatic classification of single facial images, IEEE Transactions on Pattern Analysis and Machine Intelligence 21 (12) (1999) 1357–1362.

[4] G. Donato, M. Bartlett, J. Hager, P. Ekman, T.Sejnowski, Classifying facial actions, IEEE Transactions on Pattern Analysis and Machine Intelligence 21 (10) (1999) 974–989.

[5] M.S. Bartlett, G. Littlewort, M. Frank, C. Lainscsek,I. Fasel, J. Movellan, Recognizing facial expression: machine learning and application to spontaneous behavior, in: IEEE Conference on Computer Vision andPattern Recognition (CVPR), 2005.

[6] Z. Zhang, M.J. Lyons, M. Schuster, S. Akamatsu,Comparison between geometry-based and Gabor-wavelets-based facial expression recognition using multi-layer perceptron, in: IEEE International Conference on Automatic Face & Gesture Recognition (FG), 1998.

[7] Y. Tian, Evaluation of face resolution for expression analysis, in: CVPR Workshop on Face Processing in Video, 2004.

[8] C. Shan, S. Gong, and P.W. McOwan, “Facial Expression Recognition based on Local Binary Patterns: A Comprehensive Study,” Image Vision Computing., vol. 27, no. 6, May 2009, pp. 803-816.

本文标签： RecognitionbasedFacialExpressionlocal