Tunneled Latent Variables Method for Facial Action Unit Tracking
针对表情动作单元跟踪的隧道隐变量法

A nonlinear data reduction process with limited training examples usually results in a latent variables space with some unpleasant disconnections which may cause a tracking procedure to fail at local minimums. This paper presents a novel method to resolve the problem. It is used to track facial action units in the video and the experiment results are encouraging.