郑戍华, 陈梦心, 王向周, 弓雪雅. 一种基于光流双输入网络的微表情顶点帧检测方法[J]. 北京理工大学学报自然版, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 引用本文: 郑戍华, 陈梦心, 王向周, 弓雪雅. 一种基于光流双输入网络的微表情顶点帧检测方法[J]. 北京理工大学学报自然版, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 ZHENG Shuhua, CHEN Mengxin, WANG Xiangzhou, GONG Xueya. A Micro-Expression Apex Frame Spotting Method Based on Optical-Flow-Dual-Input Network[J]. Transactions of Beijing institute of Technology, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 Citation: ZHENG Shuhua, CHEN Mengxin, WANG Xiangzhou, GONG Xueya. A Micro-Expression Apex Frame Spotting Method Based on Optical-Flow-Dual-Input Network[J]. Transactions of Beijing institute of Technology , 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 郑戍华, 陈梦心, 王向周, 弓雪雅. 一种基于光流双输入网络的微表情顶点帧检测方法[J]. 北京理工大学学报自然版, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 引用本文: 郑戍华, 陈梦心, 王向周, 弓雪雅. 一种基于光流双输入网络的微表情顶点帧检测方法[J]. 北京理工大学学报自然版, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 ZHENG Shuhua, CHEN Mengxin, WANG Xiangzhou, GONG Xueya. A Micro-Expression Apex Frame Spotting Method Based on Optical-Flow-Dual-Input Network[J]. Transactions of Beijing institute of Technology, 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 Citation: ZHENG Shuhua, CHEN Mengxin, WANG Xiangzhou, GONG Xueya. A Micro-Expression Apex Frame Spotting Method Based on Optical-Flow-Dual-Input Network[J]. Transactions of Beijing institute of Technology , 2022, 42(7): 749-754. doi: 10.15918/j.tbit1001-0645.2021.135 微表情顶点帧蕴含着丰富的微表情信息，为了准确地检测出微表情顶点帧，本文提出了一种基于光流特征的神经网络分类，并利用先验知识规则进行取舍的检测方法. 该方法针对固定滑窗大小内的图像进行光流信息提取，利用双输入特征提取网络对 x , y 方向的光流信息进行时空特征提取，并进行分类，经根据微表情先验知识所设计的取舍规则后处理后，改善了检测准确度. 实验结果表明，在数据集CASMEⅡ上测试，顶点定位率（apex spotting rate，ASR）指标达到了0.945， F ₁ -score指标达到了0.925.

微表情顶点帧 /  双输入网络 /  分类后处理 Abstract: Micro-expression apex frame contains abundant micro-expression information. In order to spot the apex frame accurately, a neural network classification was proposed based on optical flow characteristics. Taking prior knowledge as rules, a detection method was designed to realize micro-expression apex frame spotting. Firstly, optical flow information was extracted from the image in a fixed size sliding window. And then, the spatial and temporal features of optical flow information in x and y directions was extracted and classified based on dual input network. Finally, according to the trade-off rules based on prior knowledge of micro expression, a post-processing was carried out to improve the detection accuracy. The experimental results on data set CASMEⅡtesting show that the apex spotting rate (ASR) and F ₁ -score can reach up to 0.945 and 0.925 respectively.

Key words: micro-expression apex frame /  dual input network /  classification post processing