Towards Automatic Recognition of Attitudes: Prosodic Analysis of Video Blogs

Madzlan, Noor Alhusna; Han, Jian; Bonin, Francesca; Campbell, Nick

doi:10.21437/speechprosody.2014-6

Cited by 9 publications

(5 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results obtained using MFCC (228 features) are set as baseline in this study. Previous studies [12,11,10,15] do not evaluate the MFCC features for attitude recognition and use a large number of features (even more than number of instances [15]) which may result in over-fitting of machine learning models due to curse of dimensionality. However, in this study we used MFCC features for the classification task and also reduced the dimensionality of feature set using PCA as well as the proposed new method (AFT).…”

Section: Resultsmentioning

confidence: 99%

“…However, they did not perform fusion of features. In a different study [11], authors analyzed prosodic features of vlogger and found that these features (F0, voice quality and intensity) are correlated with a vlogger attitude, while in [12] they analyzed audio-visual features of vloggers for their attitude recognition. In all of the above studies, authors extracted the acoustic features using statistical functions (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Active Feature Transformation Method for Attitude Recognition of Video Bloggers

et al. 2018

View full text Add to dashboard Cite

Video blogging is a form of unidirectional communication where a video blogger expresses his/her opinion about different issues. The success of a video blog is measured using metrics like the number of views and comments by online viewers. Researchers have highlighted the importance of non-verbal behaviours (e.g. attitudes) in the context of video blogging and showed that it correlates with the level of attention (number of views) gained by a video blog. Therefore, an automatic attitude recognition system can help potential video bloggers to train their attitudes. It can also be useful in developing video blogs summarization and searching tools. This study proposes a novel Active Feature Transformation (AFT) method for automatic recognition of attitudes (a form of non-verbal behaviour) in video blogs. The proposed method transforms the Melfrequency Cepstral Coefficient (MFCC) features for the classification task. The Principal Component Analysis (PCA) transformation is also used for comparison. Our results show that AFT outperforms PCA in terms of accuracy and dimensionality reduction for attitude recognition using linear discrimination analysis, 1-nearest neighbour and decision tree classifiers.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

An Active Feature Transformation Method for Attitude Recognition of Video Bloggers

et al. 2018

View full text Add to dashboard Cite

show abstract

“…They are affected by certain biases such as reputation and self-presentation [9] as an individual is likely to answer the questions in a way that maintains the image they wish to portray to others or that they determine to be more socially desirable [10]. Since previous psychological studies [11]- [14] frequently show that personality traits can be reflected by human nonverbal behaviours, most existing personality computing approaches aim to directly recognise apparent personality traits from the target subject's audio [15]- [17], visual [18]- [20] or audio-visual behaviours [21], [22]. There is evidence suggesting that an individual's response to certain situations largely depends on their personalities [23].…”

Section: Introductionmentioning

confidence: 99%

A Framework for Automatic Personality Recognition in Dyadic Interactions

Dodd,

Song,

Gunes

2023

2023 11th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW)

View full text Add to dashboard Cite

Research has shown that the way in which an individual interacts with others contains vital cues for recognising their real personality traits. The ability to recognise and adapt to the personality of users is key to developing more intelligent social robots, especially in real-world scenarios. However, most methods for personality recognition focus on apparent personality recognition of individuals in isolated settings. In this work, we propose the first multi-modal framework for human behaviour primitives-based automatic real personality recognition in dyadic interactions. It leverages the use of the spectral representations of behavioural primitives to exploit the temporal nature of the data whilst retaining as much vital information pertaining to personality as possible. We experiment on a range of standard fusion methods to evaluate their effectiveness at combining information from multiple modalities and both interactants in a dyadic interaction. At the multi-subject level, our attention-based fusion approach using a multimodal transformer enabled with cross-subject attention was the most successful. The experimental results show that our approach improved on the previous stateof-the-art on the UDIVA dataset by up to 46%.

show abstract

“…Before describing the followed approach, we provide a brief literature review on automatic personality trait recognition. In the past, various approaches have been used for recognizing apparent personality traits from different modalities such as audio [4,5], text [6][7][8] and visual information [9,10]. As in other recognition problems, multimodal systems are also investigated to improve robustness of prediction [11][12][13][14].…”

Section: Introduction and Related Workmentioning

confidence: 99%

“…In recent approaches to personality impressions classification, Support Vector Machines (SVM) [38] have been widely used [5,8,12,14]. Recently, a learning approach called Extreme Learning Machines (ELM) that is similar to SVMs but providing faster learning schemes has become popular [39].…”

Section: Introduction and Related Workmentioning

confidence: 99%

Combining Deep Facial and Ambient Features for First Impression Estimation

Gürpınar

Kaya

Salah

2016

Lecture Notes in Computer Science

View full text Add to dashboard Cite

First impressions influence the behavior of people towards a newly encountered person or a human-like agent. Apart from the physical characteristics of the encountered face, the emotional expressions displayed on it, as well as ambient information affect these impressions. In this work, we propose an approach to predict the first impressions people will have for a given video depicting a face within a context. We employ pre-trained Deep Convolutional Neural Networks to extract facial expressions, as well as ambient information. After video modeling, visual features that represent facial expression and scene are combined and fed to a Kernel Extreme Learning Machine regressor. The proposed system is evaluated on the ChaLearn Challenge Dataset on First Impression Recognition, where the classification target is the "Big Five" personality trait labels for each video. Our system achieved an accuracy of 90.94 % on the sequestered test set, 0.36 % points below the top system in the competition.

show abstract

Towards Automatic Recognition of Attitudes: Prosodic Analysis of Video Blogs

Cited by 9 publications

References 1 publication

An Active Feature Transformation Method for Attitude Recognition of Video Bloggers

An Active Feature Transformation Method for Attitude Recognition of Video Bloggers

A Framework for Automatic Personality Recognition in Dyadic Interactions

Combining Deep Facial and Ambient Features for First Impression Estimation

Contact Info

Product

Resources

About