Abstract
This paper is concerned with video-based facial expression recognition frequently used in conjunction with HRI (Human-Robot Interaction) that can naturally interact between human and robot. For this purpose, we design a 3D-CNN(3D Convolutional Neural Networks) by augmenting dimensionality reduction methods such as PCA(Principal Component Analysis) and TMPCA(Tensor-based Multilinear Principal Component Analysis) to recognize simultaneously the successive frames with facial expression images obtained through video camera. The 3D-CNN can achieve some degree of shift and deformation invariance using local receptive fields and spatial subsampling through dimensionality reduction of redundant CNN’s output. The experimental results on video-based facial expression database reveal that the presented method shows a good performance in comparison to the conventional methods such as PCA and TMPCA.
Highlights
HRI (Human-Robot Interaction) is a critical technology as evaluating, developing and designing interactional environments for intelligent system to make cognitive and emotional interaction through some communicational channels between human and robot
We especially focus on video-based facial expression recognition technique
Methods referred above are a little difficult to use for video-based facial expression recognition, including successive frames with facial expression images
Summary
HRI (Human-Robot Interaction) is a critical technology as evaluating, developing and designing interactional environments for intelligent system to make cognitive and emotional interaction through some communicational channels between human and robot. It is for synthetically understanding a user’s intention and responding [1]-[5]. That means we need to study facial expression recognition using a stationary image and video having time base [18][19]. Methods referred above are a little difficult to use for video-based facial expression recognition, including successive frames with facial expression images.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have