Local binary patterns (LBPs) are a popular method in texture characterization. LBPs provide remarkable results under taxing situations, such as scale-, illumination-, and rotation-invariant recognition. Human emotion recognition through facial images is gaining attention, especially in human-computer interfaces.

In this paper, Moore and Bowden use LBPs for human facial expression recognition. They incorporate images with different facial expressions such as joy, surprise, fear, sadness, anger, and disgust, with varying yaw angles from 0 degrees to 90 degrees. Using a multi-pie dataset, they present extensive experimental results.

This technically sound paper presents some interesting findings. It is easy to follow, even for a reader who has only a modest knowledge of pattern classification methods. This paper will be useful to researchers working in pattern recognition, especially with regard to human-computer interfaces.