Fusion of classifier predictions for audio-visual emotion recognition

Noroozi, Fatemeh; Marjanović, Marina; Njeguš, Angelina; Escalera, Sérgio; Anbarjafari, Gholamreza

doi:10.1109/icpr.2016.7899608

Cited by 18 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [136] the authors focused on uncovering the emotional effect on the interrelation between speech and body gestures. They used prosody and mel-frequency cepstral coefficient (MFCC) [137], [138] for speech with three types of body gestures: head motion, lower and upper body motions to study the relationship between the two communication channels affected by the emotional state. Additionally, they proposed a framework for modeling the dynamics of speech-gesture interaction.…”

Section: Emotion Recognitionmentioning

confidence: 99%

Survey on Emotional Body Gesture Recognition

Noroozi

Corneanu

Kamińska

et al. 2021

IEEE Trans. Affective Comput.

Self Cite

267

134

View full text Add to dashboard Cite

Automatic emotion recognition has become a trending research topic in the past decade. While works based on facial expressions or speech abound recognizing affect from body gestures remains a less explored topic. We present a new comprehensive survey hoping to boost research in the field. We first introduce emotional body gestures as a component of what is commonly known as "body language" and comment general aspects as gender differences and culture dependence. We then define a complete framework for automatic emotional body gesture recognition. We introduce person detection and comment static and dynamic body pose estimation methods both in RGB and 3D. We then comment the recent literature related to representation learning and emotion recognition from images of emotionally expressive gestures. We also discuss multi-modal approaches that combine speech or face with body gestures for improved emotion recognition. While pre-processing methodologies (e.g. human detection and pose estimation) are nowadays mature technologies fully developed for robust large scale analysis, we show that for emotion recognition the quantity of labelled data is scarce, there is no agreement on clearly defined output spaces and the representations are shallow and largely based on naive geometrical representations.

show abstract

Section: Emotion Recognitionmentioning

confidence: 99%

Survey on Emotional Body Gesture Recognition

Noroozi

Corneanu

Kamińska

et al. 2021

IEEE Trans. Affective Comput.

Self Cite

267

134

View full text Add to dashboard Cite

show abstract

“…In the early fusion model, features from both modalities are concatenated and represented as an input to a three-layer multiperceptron classifier using a simple cross-entropy loss. In the late fusion model, the decisions (i.e, the confidence outputs from each modality) are used as input to a fully connected layer for prediction in a stacked manner [79]. Table 2 and 3 compare the accuracy on emotion recognition between convergence, enhancement, and synchrony models with the above two CNN baselines.…”

Section: Effectiveness In Multisensory Emotion Recognitionmentioning

confidence: 99%

Investigating Multisensory Integration in Emotion Recognition Through Bio-Inspired Computational Models

Benssassi

2023

IEEE Trans. Affective Comput.

View full text Add to dashboard Cite

“…To create machines that can recognize emotions, a great body of research has been dedicated to investigating the neural correlates of emotions. In these efforts, emotions have been elicited by different pictorial [1], musical [2][3][4] and video [5][6][7] stimuli. Music listening comprises a variety of psychological processes, e.g.…”

Section: Introductionmentioning

confidence: 99%