The AffectMove 2021 Challenge - Affect Recognition from Naturalistic Movement Data

Olugbade, Temitayo; Sagoleo, Roberto; Ghisio, Simone; Gold, Nicolas; Williams, Amanda C de C; Gelder, Béatrice de; Camurri, Antonio; Volpe, Gualtiero; Bianchi-Berthouze, Nadia

doi:10.1109/aciiw52867.2021.9666322

Cited by 2 publications

(7 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The AffectMove 2021 Challenge [4] is divided into three tasks. Task 1 of the competition promotes protective behavior detection based on subjects with chronic musculoskeletal pain from the Emopain dataset [28].…”

Section: ) Emopain 2021 Datasetmentioning

confidence: 99%

Pain Recognition With Physiological Signals Using Multi-Level Context Information

et al. 2023

View full text Add to dashboard Cite

Automatic pain recognition is essential in healthcare. In previous studies, automatic pain recognition methods preferentially apply the features extracted from physiological signals for conventional models. This method provides good performance but mainly relies on medical expertise for feature extraction of physiological signals. This paper presents a deep learning approach based on physiological signals that have the role of both feature extraction and classification, regardless of medical expertise. We propose multi-level context information for each physiological signal discriminating between pain and painlessness. Our experimental results prove that multi-level context information has more significant performances than uni-level context information based on Part A of the BioVid Heat Pain database and the Emopain 2021 dataset. For Part A of the BioVid Heat Pain database, our experimental results for pain recognition tasks include: Pain 0 and Pain 1, Pain 0 and Pain 2, Pain 0 and Pain 3, and Pain 0 and Pain 4. In the classification task between Pain 0 and Pain 4, the results achieve an average accuracy of 84.8 ± 13.3% for 87 subjects and 87.8 ± 11.4% for 67 subjects in a Leave-One-Subject-Out cross-validation evaluation. The proposed method adopts the ability of deep learning to outperform conventional methods on physiological signals. INDEX TERMS pain recognition, physiological signals, context vector, attention module, deep learning.

show abstract

Section: ) Emopain 2021 Datasetmentioning

confidence: 99%

Pain Recognition With Physiological Signals Using Multi-Level Context Information

et al. 2023

View full text Add to dashboard Cite

show abstract

“…In the following we index by t ∈ 1, 180 the different timesteps composing a given input to classify. 1 For a timestep t, we also denote respectively the 4×1D EMG input values and the 17×3D MoCap input values by x t EM G ∈ R 4 , and x t M OC ∈ R 51 respectively. In this section x t = (x t EM G , x t M OC ) is the concatenation of the two vectors x t EM G and x t M OC at time t (x t ∈ R 55 ).…”

Section: B Transformer-based Architecturesmentioning

confidence: 99%

“…In this section x t = (x t EM G , x t M OC ) is the concatenation of the two vectors x t EM G and x t M OC at time t (x t ∈ R 55 ). The input provided to the transformer model was X = [x (1) , . .…”

Section: B Transformer-based Architecturesmentioning

confidence: 99%

“…. (B (1) (X))). A linear transformation is first applied element-wise to each of the timestep entries of X ∈ R 55×180 to obtain the R d×180 input that will be given to the first transformer block, i.e.…”

Section: B Transformer-based Architecturesmentioning

confidence: 99%

“…The 2021 AffectMove challenge [1] organised in the context of the EnTimeMent H2020 project and collocated, as a workshop, with the ACII 20201 conference, offered the opportunity of participating in three challenge tasks that involve the detection of characteristics of human motion in multimodal and unimodal settings:…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Comparison of Deep Learning Approaches for Protective Behaviour Detection Under Class Imbalance from MoCap and EMG data

Radouane¹,

Tchechmedjiev²,

Xu³

et al. 2021

2021 9th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos (ACIIW)

View full text Add to dashboard Cite

The AffecMove challenge organised in the context of the H2020 EnTimeMent project offers three tasks of movement classification in realistic settings and use-cases. Our team, from the EuroMov DHM laboratory participated in Task 1, for protective behaviour (against pain) detection from motion capture data and EMG, in patients suffering from pain-inducing muskuloskeletal disorders. We implemented two simple baseline systems, one LSTM system with pre-training (NTU-60) and a Transformer. We also adapted PA-ResGCN a Graph Convolutional Network for skeleton-based action classification showing state-of-the-art (SOTA) performance to protective behaviour detection, augmented with strategies to handle class-imbalance. For PA-ResGCN-N51 we explored naïve fusion strategies with an EMG-only convolutional neural network that didn't improve the overall performance. Unsurprisingly, the best performing system was PA-ResGCN-N51 (w/o EMG) with a F1 score of 53.36% on the test set for the minority class (MCC 0.4247). The Transformer baseline (MoCap + EMG) came second at 41.05% F1 test performance (MCC 0.3523) and the LSTM baseline third at 31.16% F1 (MCC 0.1763). On the validation set the LSTM showed performance comparable to PA-ResGCN, we hypothesize that the LSTM over-fitted on the validation set that wasn't very representative of the train/test distribution.

show abstract

The AffectMove 2021 Challenge - Affect Recognition from Naturalistic Movement Data

Cited by 2 publications

References 33 publications

Pain Recognition With Physiological Signals Using Multi-Level Context Information

Pain Recognition With Physiological Signals Using Multi-Level Context Information

Comparison of Deep Learning Approaches for Protective Behaviour Detection Under Class Imbalance from MoCap and EMG data

Contact Info

Product

Resources

About