Comparison of Deep Learning Approaches for Protective Behaviour Detection Under Class Imbalance from MoCap and EMG data - Archive ouverte HAL Access content directly
Conference Papers Year : 2021

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

Abstract

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.
Fichier principal
Vignette du fichier
AffecMov_Challenge_EuroMov_DHM_Deep (3).pdf (965.13 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03523502 , version 1 (14-01-2022)

Identifiers

Cite

Karim Radouane, Andon Tchechmedjiev, Binbin Xu, Sebastien Harispe. Comparison of Deep Learning Approaches for Protective Behaviour Detection Under Class Imbalance from MoCap and EMG data. ACIIW 2021 - 9th International Conference on Affective Computing and Intelligent Interaction Workshops and Demos, Sep 2021, Nara, Japan. pp.01-08, ⟨10.1109/ACIIW52867.2021.9666417⟩. ⟨hal-03523502⟩
384 View
173 Download

Altmetric

Share

Gmail Facebook X LinkedIn More