Expression Recognition in Videos Using a Weighted Component-Based Feature Descriptor

Huang, Xiaohua; Zhao, Guoying; Pietikäinen, Matti; Zheng, Wenming

doi:10.1007/978-3-642-21227-7_53

Cited by 11 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[35][36][37][38] More recent works have investigated the problem of expression analysis of videos. [39][40][41] To deal with such a dynamic process, the methods must be capable of effectively considering temporal alignment and semantic representation. 40 In order to promote and improve the development of automatic expression recognition approaches, several challenges 42,43 and datasets [44][45][46][47][48][49][50][51] have been created which aim to establish a common platform for creating and validating expression recognition methods in both controlled and realworld conditions.…”

Section: Introductionmentioning

confidence: 99%

Effects of cultural characteristics on building an emotion classifier through facial expression analysis

Silva

Pedrini

2015

J. Electron. Imaging

View full text Add to dashboard Cite

Facial expressions are an important demonstration of humanity's humors and emotions. Algorithms capable of recognizing facial expressions and associating them with emotions were developed and employed to compare the expressions that different cultural groups use to show their emotions. Static pictures of predominantly occidental and oriental subjects from public datasets were used to train machine learning algorithms, whereas local binary patterns, histogram of oriented gradients (HOGs), and Gabor filters were employed to describe the facial expressions for six different basic emotions. The most consistent combination, formed by the association of HOG filter and support vector machines, was then used to classify the other cultural group: there was a strong drop in accuracy, meaning that the subtle differences of facial expressions of each culture affected the classifier performance. Finally, a classifier was trained with images from both occidental and oriental subjects and its accuracy was higher on multicultural data, evidencing the need of a multicultural training set to build an efficient classifier.

show abstract

Section: Introductionmentioning

confidence: 99%

Effects of cultural characteristics on building an emotion classifier through facial expression analysis

Silva

Pedrini

2015

J. Electron. Imaging

View full text Add to dashboard Cite

show abstract

“…Happiness and surprise all keep good performance while the recognition accuracy of other facial expressions increase significantly. Then, we compare the performance of our algorithm with the some state-of-the-art facial expression recognition approaches, including active appearance models (AAM) tracked similarity-normalized shape (SPTS) and canonical appearance (CAPP) features with linear SVM [21], constrained local models (CLM) tracked SPTS and SAPP features with linear SVM [22], emotion avatar image (EAI) and local phase quantization (LPQ) features with linear SVM [23] and weighted component-based feature descriptor algorithm [24]. Table. IV gives the classification accuracy of the state-of-theart facial expression recognition algorithms and the proposed method.…”

Section: B Experiments On the Extended Cohn-kanade Datasetmentioning

confidence: 99%

LogDet Divergence-Based Metric Learning With Triplet Constraints and Its Applications

Mei

Liu

Karimi

et al. 2014

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

How to select and weigh features has always been a difficult problem in many image processing and pattern recognition applications. A data-dependent distance measure can address this problem to a certain extent, and therefore an accurate and efficient metric learning becomes necessary. In this paper, we propose a LogDet divergence-based metric learning with triplet constraints (LDMLT) approach, which can learn Mahalanobis distance metric accurately and efficiently. First of all, we demonstrate the good properties of triplet constraints and apply it in LogDet divergence-based metric learning model. Then, to deal with high-dimensional data, we apply a compressed representation method to learn, store, and evaluate Mahalanobis matrix efficiently. Besides, a dynamic triplets building strategy is proposed to build a feedback from the obtained Mahalanobis matrix to the triplet constraints, which can further improve the LDMLT algorithm. Furthermore, the proposed method is applied to various applications, including pattern recognition, facial expression recognition, and image retrieval. The results demonstrate the improved performance of the proposed approach.

show abstract

“…The main idea of the Fisher criterion is to learn the weights of each region based on keeping the within-class scatter as small as possible and between-class scatter as large as possible [16,17]. For the C class problem, the similarities of the different samples from the same class form the within-class scatter, while the difference samples from different classes form the between-class scatter.…”

Section: Weights Learned Based On Fisher Criterionmentioning

confidence: 99%

Dynamic biometric identification from multiple views using the GLBP-TOP method

Wang

Shen

Chen

et al. 2014

Bio-Medical Materials and Engineering

View full text Add to dashboard Cite

To realize effective and rapid dynamic biometric identification with low computational complexity, a video-based facial texture program that extracts local binary patterns from three orthogonal planes in the frequency domain of the Gabor transform (GLBP-TOP) was proposed. Firstly, each normalized face was transformed by Gabor wavelet to get the enhanced Gabor magnitude map, and then the LBP-TOP operator was applied to the maps to extract video texture. Finally, weighted Chi square statistics based on the Fisher Criterion were used to realize the identification. The proposed algorithm was proved effective through the biometric experiments using the Honda/UCSD database, and was robust against changes of illumination and expressions.

show abstract

Expression Recognition in Videos Using a Weighted Component-Based Feature Descriptor

Cited by 11 publications

References 25 publications

Effects of cultural characteristics on building an emotion classifier through facial expression analysis

Effects of cultural characteristics on building an emotion classifier through facial expression analysis

LogDet Divergence-Based Metric Learning With Triplet Constraints and Its Applications

Dynamic biometric identification from multiple views using the GLBP-TOP method

Contact Info

Product

Resources

About