Marwa Mahmoud

Automatic behavior descriptors for psychological disorder analysis

We investigate the capabilities of automatic nonverbal behavior descriptors to identify indicators of psychological disorders such as depression, anxiety, and post-traumatic stress disorder. We seek to confirm and enrich present state of the art, predominantly based on qualitative manual annotations, with automatic quantitative behavior descriptors. In this paper, we propose four nonverbal behavior descriptors that can be automatically estimated from visual signals. We introduce a new dataset called the Distress Assessment Interview Corpus (DAIC) which includes 167 dyadic interactions between a confederate interviewer and a paid participant. Our evaluation on this dataset shows correlation of our automatic behavior descriptors with specific psychological disorders as well as a generic distress measure. Our analysis also includes a deeper study of self-adaptor and fidgeting behaviors based on detailed annotations of where these behaviors occur.

Cross-dataset learning and person-specific normalisation for automatic Action Unit detection

Automatic detection of Facial Action Units (AUs) is crucial for facial analysis systems. Due to the large individual differences, performance of AU classifiers depends largely on training data and the ability to estimate facial expressions of a neutral face. In this paper, we present a real-time Facial Action Unit intensity estimation and occurrence detection system based on appearance (Histograms of Oriented Gradients) and geometry features (shape parameters and landmark locations). Our experiments show the benefits of using additional labelled data from different datasets, which demonstrates the generalisability of our approach. This holds both when training for a specific dataset or when a generic model is needed. We also demonstrate the benefits of using a simple and efficient median based feature normalisation technique that accounts for person-specific neutral expressions. Finally, we show that our results outperform the FERA 2015 baselines in all three challenge tasks - AU occurrence detection, fully automatic AU intensity and pre-segmented AU intensity estimation.

3D corpus of spontaneous complex mental states

Hand-over-face gestures, a subset of emotional body language, are overlooked by automatic affect inference systems. We propose the use of hand-over-face gestures as a novel affect cue for automatic inference of cognitive mental states. Moreover, affect recognition systems rely on the existence of publicly available datasets, often the approach is only as good as the data. We present the collection and annotation methodology of a 3D multimodal corpus of 108 audio/video segments of natural complex mental states. The corpus includes spontaneous facial expressions and hand gestures labelled using crowd-sourcing and is publicly available.

Real-time inference of mental states from facial expressions and upper body gestures

We present a real-time system for detecting facial action units and inferring emotional states from head and shoulder gestures and facial expressions. The dynamic system uses three levels of inference on progressively longer time scales. Firstly, facial action units and head orientation are identified from 22 feature points and Gabor filters. Secondly, Hidden Markov Models are used to classify sequences of actions into head and shoulder gestures. Finally, a multi level Dynamic Bayesian Network is used to model the unfolding emotional state based on probabilities of different gestures. The most probable state over a given video clip is chosen as the label for that clip. The average F1 score for 12 action units (AUs 1, 2, 4, 6, 7, 10, 12, 15, 17, 18, 25, 26), labelled on a frame by frame basis, was 0.461. The average classification rate for five emotional states (anger, fear, joy, relief, sadness) was 0.440. Sadness had the greatest rate, 0.64, anger the smallest, 0.11.

Automatic Audiovisual Behavior Descriptors for Psychological Disorder Analysis

Abstract We investigate the capabilities of automatic audiovisual nonverbal behavior descriptors to identify indicators of psychological disorders such as depression, anxiety, and post-traumatic stress disorder. Due to strong correlations between these disordersas measured with standard self-assessment questionnaires in this study, we focus our investigations in particular on a generic distress measure as identified using factor analysis. Within this work, we seek to confirm and enrich present state of the art, predominantly based on qualitative manual annotations, with automatic quantitative behavior descriptors. We propose a number of nonverbal behavior descriptors that can be automatically estimated from audiovisual signals. Such automatic behavior descriptors could be used to support healthcare providers with quantified and objective observations that could ultimately improve clinical assessment. We evaluate our work on the dataset called the Distress Assessment Interview Corpus (DAIC) which comprises dyadic interactions between a confederate interviewer and a paid participant. Our evaluation on this dataset shows correlation of our automatic behavior descriptors with the derived general distress measure. Our analysis also includes a deeper study of self-adaptor and fidgeting behaviors based on detailed annotations of where these behaviors occur.

Interpreting hand-over-face gestures

People often hold their hands near their faces as a gesture in natural conversation, which can interfere with affective inference from facial expressions. However, these gestures are valuable as an additional channel for multi-modal inference. We analyse hand-over-face gestures in a corpus of naturalistic labelled expressions and propose the use of those gestures as a novel affect cue for automatic inference of cognitive mental states. We define three hand cues for encoding hand-over-face gestures, namely hand shape, hand action and facial region occluded, serving as a first step in automating the interpretation process.

Automatic Detection of Naturalistic Hand-over-Face Gesture Descriptors

One of the main factors that limit the accuracy of facial analysis systems is hand occlusion. As the face becomes occluded, facial features are either lost, corrupted or erroneously detected. Hand-over-face occlusions are considered not only very common but also very challenging to handle. Moreover, there is empirical evidence that some of these hand-over-face gestures serve as cues for recognition of cognitive mental states. In this paper, we detect hand-over-face occlusions and classify hand-over-face gesture descriptors in videos of natural expressions using multi-modal fusion of different state-of-the-art spatial and spatio-temporal features. We show experimentally that we can successfully detect face occlusions with an accuracy of 83%. We also demonstrate that we can classify gesture descriptors (hand shape, hand action and facial region occluded) significantly higher than a naive baseline. To our knowledge, this work is the first attempt to automatically detect and classify hand-over-face gestures in natural expressions.

Automatic multimodal descriptors of rhythmic body movement

Prolonged durations of rhythmic body gestures were proved to be correlated with different types of psychological disorders. To-date, there is no automatic descriptor that can robustly detect those behaviours. In this paper, we propose a cyclic gestures descriptor that can detect and localise rhythmic body movements by taking advantage of both colour and depth modalities. We show experimentally how our rhythmic descriptor can successfully localise the rhythmic gestures as: hands fidgeting, legs fidgeting or rocking, significantly higher than the majority vote classification baseline. Our experiments also demonstrate the importance of fusing both modalities, with a significant increase in performance when compared to individual modalities.

Towards Communicative Face Occlusions: Machine Detection of Hand-over-Face Gestures

Emotional body language constitutes an important channel of non-verbal information. Of this large set, hand-over-face gestures are treated as noise because they occlude facial expressions. In this paper, we propose an alternative facial processing framework where face occlusions instead of being removed, are detected, localized and eventually classified into communicative gestures. We present a video corpus of hand-over-face gestures and describe a multi-stage methodology for detecting and localizing these gestures. For pre-processing, we show that force fields form a better representation of images compared to edge detectors. For feature extraction, detection and localization, we show that Local Binary Patterns outperform Gabor filters in accuracy and speed. Our methodology yields an average detection rate of 97%, is robust to changes in facial expressions, hand shapes, and limited head motion, and preliminary testing with spontaneous videos suggests that it may generalize successfully to naturally evoked videos.

Decoupling facial expressions and head motions in complex emotions

Perception of emotion through facial expressions and head motion is of interest to both psychology and affective computing researchers. However, very little is known about the importance of each modality individually, as they are often treated together rather than separately. We present a study which isolates the effect of head motion from facial expression in the perception of complex emotions in videos. We demonstrate that head motions carry emotional information that is complementary rather than redundant to the emotion content in facial expressions. Finally, we show that emotional expressivity in head motion is not limited to nods and shakes and that additional gestures (such as head tilts, raises and general amount of motion) could be beneficial to automated recognition systems.