Evolving robot empathy towards humans with motor disabilities through artificial pain generation

Anshar, Muh; Williams, Mary-Anne

doi:10.3934/neuroscience.2018.1.56

Cited by 1 publication

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers in the field of human-robot interaction (HRI) have envisioned the use of automatic empathy prediction systems to improve HRI [13], [5]. In this regard, previous approaches analysing empathy from audio-and video-based features have resulted in promising findings [39] due to the richness in emotional content of both modalities.…”

Section: Introductionmentioning

confidence: 99%

Performance Analysis of Unimodal and Multimodal Models in Valence-Based Empathy Recognition

Mallol-Ragolta

Schmitt

Baird

et al. 2019

2019 14th IEEE International Conference on Automatic Face &Amp; Gesture Recognition (FG 2019)

View full text Add to dashboard Cite

The human ability to empathise is a core aspect of successful interpersonal relationships. In this regard, humanrobot interaction can be improved through the automatic perception of empathy, among other human attributes, allowing robots to affectively adapt their actions to interactants' feelings in any given situation. This paper presents our contribution to the generalised track of the One-Minute Gradual (OMG) Empathy Prediction Challenge by describing our approach to predict a listener's valence during semi-scripted actor-listener interactions. We extract visual and acoustic features from the interactions and feed them into a bidirectional long shortterm memory network to capture the time-dependencies of the valence-based empathy during the interactions. Generalised and personalised unimodal and multimodal valence-based empathy models are then trained to assess the impact of each modality on the system performance. Furthermore, we analyse if intrasubject dependencies on empathy perception affect the system performance. We assess the models by computing the concordance correlation coefficient (CCC) between the predicted and self-annotated valence scores. The results support the suitability of employing multimodal data to recognise participants' valence-based empathy during the interactions, and highlight the subject-dependency of empathy. In particular, we obtained our best result with a personalised multimodal model, which achieved a CCC of 0.11 on the test set.

show abstract

Section: Introductionmentioning

confidence: 99%