基于三维卷积神经网络和峰值帧光流的微表情识别算法

针对现有微表情识别技术未能有效利用峰值帧前后时间空间特征的缺点,文中提出基于三维卷积神经网络和峰值帧光流的微表情识别算法.首先,提取峰值帧前后相邻帧间的光流场,在保留微表情重要时间、空间信息的同时,去除冗余信息,减少计算量.然后,利用三维卷积神经网络,从光流场中提取增强的时空特征,实现微表情的分类识别.最后,通过在3个微表情数据库上的对比实验证实文中算法准确度较高.

Micro-Expression Recognition Algorithm Based on 3D Convolutional Neural Network and Optical Flow Fields from Neighboring Frames of Apex Frame

The existing micro-expression recognition technologies cannot make full use of the spatiotemporal features near the apex frame. Aiming at this problem, a micro-expression recognition algorithm based on 3D convolutional neural network and optical flow fields from the neighboring frames of the apex frame is proposed. Firstly, the optical flow fields between the adjacent frames before and after the apex frame are extracted. The important spatiotemporal information of micro-expressions are retained while the redundant information is removed and the computation load is reduced. Then, a 3D convolutional neural network is employed to extract the enhanced spatiotemporal features from the optical flow fields and thus the classification is completed. Finally, experiments on three spontaneous micro-expression databases show the proposed algorithm produces a better accuracy.