A real time and robust facial expression recognition and imitation approach for affective human-robot interaction using Gabor filtering

Cid, Felipe; Prado, Jose; Bustos, Pablo; Núñez, Pedro

doi:10.1109/iros.2013.6696662

Cited by 33 publications

(31 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These models allow robots to interpret affective states in a similar manner as humans [75]. The most common discrete affective categories used by robots in HRI settings are disgust, sad, surprise, anger, disgust, fear, sad, happy, surprise, as well as neutral.…”

Section: Affect Models Used In Hrimentioning

confidence: 99%

“…Facial-affect classification is, then, used to estimate affect utilizing binary decision trees [58], 2) AdaBoost [102], multilayer perceptrons (MLPs) [60], support vector machines (SVMs) [103], support vector regression (SVRs) [60,76], neural networks (NNs) [60,[104][105][106][107][108], or dynamic Bayesian networks (DBNs) [75].…”

Section: Facial Affect Recognition During Hrimentioning

confidence: 99%

“…In [75], a real-time system for recognition and imitation of facial expressions was developed for the robotic head Muecas. Muecas was equipped with an onboard camera to capture images of a person's face for facial expression recognition.…”

Section: Mimicry Hrimentioning

confidence: 99%

“…An extension of the approach in [75] was presented in [199], where speech recognition from two microphones was incorporated with the facial expression recognition system in the humanoid robotic head, Muecas. Utilizing both modalities, Muecas was able to imitate human affect (happiness, sadness, anger, neutral, fear) and respond in real-time through the display of facial expressions and head pose.…”

Section: Mimicry Hrimentioning

confidence: 99%

“…Utilizing both modalities, Muecas was able to imitate human affect (happiness, sadness, anger, neutral, fear) and respond in real-time through the display of facial expressions and head pose. The facial expression recognition system was similar to that in [75] where facial expression recognition was accomplished through Gabor filtering and DBN classification. For speech recognition, the speech rate, pitch, and energy were extracted from the signals received from the audio signal.…”

Section: Mimicry Hrimentioning

confidence: 99%

See 4 more Smart Citations

A Survey of Autonomous Human Affect Detection Methods for Social Robots Engaged in Natural HRI

McColl

Hong

Hatakeyama

et al. 2015

J Intell Robot Syst

104

View full text Add to dashboard Cite

In Human-Robot Interactions (HRI), robots should be socially intelligent. They should be able to respond appropriately to human affective and social cues in order to effectively engage in bi-directional communications. Social intelligence would allow a robot to relate to, understand, and interact and share information with people in real-world humancentered environments. This survey paper presents an encompassing review of existing automated affect recognition and classification systems for social robots engaged in various HRI settings. Human-affect detection from facial expressions, body language, voice, and physiological signals are investigated, as well as from a combination of the aforementioned modes. The automated systems are described by their corresponding robotic and HRI applications, the sensors they employ, and the feature detection techniques and affect classification strategies utilized. This paper also discusses pertinent future research directions for promoting the development of socially intelligent robots capable of recognizing, classifying and responding to human affective states during real-time HRI.

show abstract

Section: Affect Models Used In Hrimentioning

confidence: 99%