Dimensional affect recognition using Continuous Conditional Random Fields

Baltrušaitis, Tadas; Banda, Ntombikayise; Robinson, Peter

doi:10.1109/fg.2013.6553785

Cited by 89 publications

(62 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CCRF have been applied in combination with SVR for the task of continuous and dimensional emotion prediction [60]. Herein, we follow the approach in [60] in using linear SVR (exactly as described above) to learn the vertex (static) features of the graphical model and ten edge (temporal) features, that is, 5 neighbor n = {1, 2, . .…”

Section: Accepted M Manuscriptmentioning

confidence: 99%

“…The two most critical parameters in the RF design, that is the number of trees T in the forest and the number of features F selected to split each node, are optimized in the range T ∈ {100, 500, 1000, 2000} and F ∈ { √ p, p/3, p/2}, respectively, where p denotes the dimensionality of the feature vector. CCRF [60] is an undirected graphical model-based discriminative framework that extends the traditional Conditional Random Fields (CRF) [68] to the case of continuous (real-valued) output. CCRF have been applied in combination with SVR for the task of continuous and dimensional emotion prediction [60].…”

Section: Accepted M Manuscriptmentioning

confidence: 99%

“…CCRF [60] is an undirected graphical model-based discriminative framework that extends the traditional Conditional Random Fields (CRF) [68] to the case of continuous (real-valued) output. CCRF have been applied in combination with SVR for the task of continuous and dimensional emotion prediction [60]. Herein, we follow the approach in [60] in using linear SVR (exactly as described above) to learn the vertex (static) features of the graphical model and ten edge (temporal) features, that is, 5 neighbor n = {1, 2, .…”

Section: Accepted M Manuscriptmentioning

confidence: 99%

“…Four classifiers that have been extensively used for temporal modeling of human behavior and affect are examined, namely Support Vector Regression (SVR) [58], Random Forests for Regression (RF) [59], Continuous Conditional Random Fields (CCRF) [60], and Long-Short Term Memory (LSTM) Neural Networks [61]. LIBSVM [62], scikit-learn [63], [60], and the CUda RecurREnt Neural Network Toolkit (CURRENNT) [64] are used to train SVR, RF, CCRF and LSTMs, respectively.…”

Section: Classifiersmentioning

confidence: 99%

“…LIBSVM [62], scikit-learn [63], [60], and the CUda RecurREnt Neural Network Toolkit (CURRENNT) [64] are used to train SVR, RF, CCRF and LSTMs, respectively. For each fold of the crossvalidation experiments, the validation set is used to optimize the classifiers in terms of Correlation (COR) for SVR, RF and CCRF (see Section 4.5), and Root Mean Squared Error (RMSE) for LSTMs 4 .…”

Section: Classifiersmentioning

confidence: 99%

See 4 more Smart Citations

The Conflict Escalation Resolution (CONFER) Database

Georgakis

Panagakis

Zafeiriou

et al. 2017

Image and Vision Computing

View full text Add to dashboard Cite

Conflict is usually defined as a high level of disagreement taking place when individuals act on incompatible goals, interests, or intentions. Research in human sciences has recognized conflict as one of the main dimensions along which an interaction is perceived and assessed. Hence, automatic estimation of conflict intensity in naturalistic conversations would be a valuable tool for the advancement of human-centered computing and the deployment of novel applications for social skills enhancement including conflict management and negotiation. However, machine analysis of conflict is still limited to just a few works, partially due to an overall lack of suitable annotated data, while it has been mostly approached as a conflict or (dis)agreement detection problem based on audio features only. In this work, we aim to overcome the aforementioned limitations by a) presenting the Conflict Escalation Resolution (CONFER) Database, a set of excerpts from audio-visual recordings of televised political debates where conflicts naturally arise, and b) reporting baseline experiments on audio-visual conflict intensity estimation. The database contains approximately 142 minutes of recordings in Greek language, split over 120 non-overlapping episodes of naturalistic conversations that involve two or three interactants. Subject-and session-independent experiments are conducted on continuous-time (frame-by-frame) estimation of real-valued conflict intensity, as opposed to binary conflict/non-conflict clas- * Corresponding author. E-mail address: christos.georgakis@imperial.ac.uk Preprint submitted to Image and Vision Computing December 20, 2016 A C C E P T E D M A N U S C R I P T ACCEPTED MANUSCRIPTsification. For the problem at hand, the efficiency of various audio and visual features and fusion of them as well as various regression frameworks is examined. Experimental results suggest that there is much room for improvement in the design and development of automated multi-modal approaches to continuous conflict analysis. The CONFER Database is publicly available for non-commercial use at http://ibug.doc.ic.ac.uk/resources/confer/.

show abstract

Section: Accepted M Manuscriptmentioning

confidence: 99%

Section: Accepted M Manuscriptmentioning

confidence: 99%

Section: Accepted M Manuscriptmentioning

confidence: 99%

Section: Classifiersmentioning

confidence: 99%

Section: Classifiersmentioning

confidence: 99%

See 3 more Smart Citations

The Conflict Escalation Resolution (CONFER) Database

Georgakis

Panagakis

Zafeiriou

et al. 2017

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

Continuous Conditional Neural Fields for Structured Regression

Baltrušaitis

Robinson

Morency

2014

Computer Vision – ECCV 2014

Self Cite

View full text Add to dashboard Cite

Abstract. An increasing number of computer vision and pattern recognition problems require structured regression techniques. Problems like human pose estimation, unsegmented action recognition, emotion prediction and facial landmark detection have temporal or spatial output dependencies that regular regression techniques do not capture. In this paper we present continuous conditional neural fields (CCNF) -a novel structured regression model that can learn non-linear input-output dependencies, and model temporal and spatial output relationships of varying length sequences. We propose two instances of our CCNF framework: Chain-CCNF for time series modelling, and Grid-CCNF for spatial relationship modelling. We evaluate our model on five public datasets spanning three different regression problems: facial landmark detection in the wild, emotion prediction in music and facial action unit recognition. Our CCNF model demonstrates state-of-the-art performance on all of the datasets used.

show abstract