Evaluation of Gabor-wavelet-based facial action unit recognition in image sequences of increasing complexity
read more
Citations
Emotional Expressions Reconsidered: Challenges to Inferring Emotion From Human Facial Movements:
Facial Expression Recognition in Image Sequences Using Geometric Deformation Features and Support Vector Machines
Facial Expression Recognition via a Boosted Deep Belief Network
Facial Expression Analysis.
Automatic, Dimensional and Continuous Emotion Recognition
References
Neural network-based face detection
Comprehensive database for facial expression analysis
Coding facial expressions with Gabor wavelets
Complete discrete 2-D Gabor transforms by neural networks for image analysis and compression
Recognizing action units for facial expression analysis
Related Papers (5)
Frequently Asked Questions (8)
Q2. What is the effect of contracting facial muscles?
Contracting the facial muscles produces changes in both the direction and magnitude of skin surface displacement, and in the appearance of permanent and transient facial features.
Q3. What is the way to detect AUs?
Most instances of AU7 are of low intensity, which change only 1 or 2 pixels in face image and cannot be extracted by geometry-feature-based method.
Q4. How many Gabor wavelet coefficients are calculated at 20 locations?
In the experiment, total 800 Gabor wavelet coefficients corresponding 5-scale and 8-orientation are calculated at 20 specific locations.
Q5. What are the two types of facial feature extraction methods?
In facial feature extraction of expression analysis, there are mainly two types of approaches: geometric featurebased methods and appearance-based methods [1, 2, 3, 5, 6, 7, 10, 11, 12, 13, 15, 17, 16, 18, 19].
Q6. What is the method for recognizing emotion-specified expressions?
Previous work suggests that the appearance-based methods (specifically Gabor wavelets) can achieve high sensitivity and specificity for emotion-specified expressions (e.g., happy, sad) [11, 20] and single AUs [5] under four conditions.
Q7. How many Gabor wavelet coefficients are calculated in 20 locations?
In their implementation, 800 Gabor wavelet coefficients are calculated in 20 locations which are automatically defined based on the geometric features in the upper face (Figure 2).
Q8. What are the facial features that are automatically detected and tracked in the image sequence?
Both permanent (e.g., brows, eyes, lips) and transient (lines and furrows) face feature changes are automatically detected and tracked in the image sequence.