Facial expression recognition based on Gabor features and sparse representation

Liu, Weifeng; Song, Caifeng; Wang, Yanjiang; Jia, Lu

doi:10.1109/icarcv.2012.6485394

Cited by 10 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For recognition procedures, Gabor Wavelet Features with five scales and eight orientations were firstly extracted [10], after which Kernel PCA [11] was performed to compress the feature data from 10240 dimensions into around 60-d with no less than 95% information preserved. Classifiers including SVM (LibSVM [12]), LDA and KNN were compared.…”

Section: B Experimental Settingmentioning

confidence: 99%

Robust One-Shot Facial Expression Recognition with Sunglasses

Jiang¹,

Huang²,

Mu³

et al. 2016

IJMLC

View full text Add to dashboard Cite

Abstract-Automatic facial expression recognition (FER) is both interesting and important in computer vision and machine intelligence. While previous FER systems often focus on learning a classifier in a controlled environment, a more practical and robust scenario is now under consideration. More specifically, traditional FER requires collecting as many as possible facial photos so that they will accurately recognize expressions no matter a particular person wears sunglasses, hats, and other accessories or not. Such requirement is however inconvenient and could impose practical difficulties for users. To alleviate this problem, a robust one-shot FER system that only requires taking one single facial photo for each expression of each user is proposed in this paper. When taking the single photo, the user is free to choose whether to wear sunglasses or not. The sunglasses can even be of different shape and of various luminous transmittance.Such one-shot recognition improves the user-friendliness of the FER system. Importantly, a novel and practical sunglasses detection and recovery approach is developed, which obtains an obvious accuracy improvement of 6.09%, 5.86% and 4.33% with state-of-the-art classifiers including Support Vector Machine (SVM), Linear Discriminate Analysis (LDA) and K-Nearest Neighbors (KNN) respectively on the modified Japanese Female Facial Expression (JAFFE) benchmark database.Index Terms-Facial expression recognition, sunglasses detection and recovery, one-shot recognition system.

show abstract

Section: B Experimental Settingmentioning

confidence: 99%

Robust One-Shot Facial Expression Recognition with Sunglasses

Jiang¹,

Huang²,

Mu³

et al. 2016

IJMLC

View full text Add to dashboard Cite

show abstract

“…Both methods are more sensitive to noise and have difficulty in describing subtle changes in facial muscle movement. Textures and edges are studied using appearance‐based methods such as local binary pattern (LBP) [10], local directional ternary pattern [11], histograms of oriented gradients (HOG) [12], and Gabor wavelet [13]. Compared to geometric‐based methods, an appearance‐based method is robust to noise and extracts appropriate discriminative features.…”

Section: Introductionmentioning

confidence: 99%

Novel deep learning model for facial expression recognition based on maximum boosted CNN and LSTM

Rajan¹,

Chenniappan²,

Devaraj

et al. 2020

IET Image Processing

View full text Add to dashboard Cite

“…w i = c f req w i f req + c size w i size + c prob w i prob (12) where c x∈{ f req,size,prob} is a constant term that is used to adjust the priority of frequency of occurrence, face size, and face probability features in the face voting process. The significant tracked faces F and corresponding tracked persons P have a weight that reaches a maximum value:…”

Section: Face Votingmentioning

confidence: 99%

“…in terms of shape, location, distance, and curvature [8][9][10]. Appearance-based approaches use local descriptors, image filters such as LBP [11], Gabor filters [12], PHOG [13], etc. to extract hand-crafted features for facial expression representation for traditional methods.…”

Section: Introductionmentioning

confidence: 99%

Context-Aware Emotion Recognition in the Wild Using Spatio-Temporal and Temporal-Pyramid Models

Kim

Yang

et al. 2021

Sensors

View full text Add to dashboard Cite

Emotion recognition plays an important role in human–computer interactions. Recent studies have focused on video emotion recognition in the wild and have run into difficulties related to occlusion, illumination, complex behavior over time, and auditory cues. State-of-the-art methods use multiple modalities, such as frame-level, spatiotemporal, and audio approaches. However, such methods have difficulties in exploiting long-term dependencies in temporal information, capturing contextual information, and integrating multi-modal information. In this paper, we introduce a multi-modal flexible system for video-based emotion recognition in the wild. Our system tracks and votes on significant faces corresponding to persons of interest in a video to classify seven basic emotions. The key contribution of this study is that it proposes the use of face feature extraction with context-aware and statistical information for emotion recognition. We also build two model architectures to effectively exploit long-term dependencies in temporal information with a temporal-pyramid model and a spatiotemporal model with “Conv2D+LSTM+3DCNN+Classify” architecture. Finally, we propose the best selection ensemble to improve the accuracy of multi-modal fusion. The best selection ensemble selects the best combination from spatiotemporal and temporal-pyramid models to achieve the best accuracy for classifying the seven basic emotions. In our experiment, we take benchmark measurement on the AFEW dataset with high accuracy.

show abstract

Facial expression recognition based on Gabor features and sparse representation

Cited by 10 publications

References 22 publications

Robust One-Shot Facial Expression Recognition with Sunglasses

Robust One-Shot Facial Expression Recognition with Sunglasses

Novel deep learning model for facial expression recognition based on maximum boosted CNN and LSTM

Context-Aware Emotion Recognition in the Wild Using Spatio-Temporal and Temporal-Pyramid Models

Contact Info

Product

Resources

About