Research on Adaptive Learning Mechanism of the Facial Expression Feature Space

在这项研究中，我们提出了一种结合模糊ART的具有附加学习能力的人脸表情识别模型。为了证明它有多有用，我们进行了一些基本的实验，用于评估目的，既检查了快乐的表情，也检查了没有面部表情的情况。实验结果表明，本文提出的方法建立的人脸表情特征空间在保持原有知识不变的情况下，能够捕捉到更多的知识。实验结果表明，该方法适用于在时间轴上具有鲁棒性的自适应人脸表情识别模型。

In this research, we proposed a facial expression recognition model with the capability of additional learning with the Fuzzy ART incorporated as well. To demonstrate how useful it can be, we conducted some basic experiments for evaluation purposes with both joyful expressions and no facial expressions examined. Results show clearly that the facial expression feature space that we created using the method we proposed was capable of capturing additional knowledge while it maintained the existing knowledge as it was without alteration. The results we obtained demonstrate that the proposed method can be useful for an adaptive facial expression recognition model that has a robustness feature along the time axis.