|本期目录/Table of Contents|

[1]陈 龙,张水平*,王海晖,等.基于多任务学习和知识图谱的面部表情识别[J].武汉工程大学学报,2021,43(06):681-688.[doi:10.19843/j.cnki.CN42-1779/TQ.202107010]
 CHEN Long,ZHANG Shuiping *,et al.Facial Expression Recognition Based on Multi-Task Learning and Knowledge Graph[J].Journal of Wuhan Institute of Technology,2021,43(06):681-688.[doi:10.19843/j.cnki.CN42-1779/TQ.202107010]
点击复制

基于多任务学习和知识图谱的面部表情识别(/HTML)
分享到:

《武汉工程大学学报》[ISSN:1674-2869/CN:42-1779/TQ]

卷:
43
期数:
2021年06期
页码:
681-688
栏目:
机电与信息工程
出版日期:
2021-12-31

文章信息/Info

Title:
Facial Expression Recognition Based on Multi-Task Learning and Knowledge Graph
文章编号:
1674 - 2869(2021)06 - 0681 - 08
作者:
陈 龙12张水平*12王海晖12陈言璞12
1. 武汉工程大学计算机科学与工程学院, 湖北 武汉 430205;2. 智能机器人湖北省重点实验室(武汉工程大学), 湖北 武汉 430205
Author(s):
CHEN Long1 2 ZHANG Shuiping *1 2 WANG Haihui1 2 CHEN Yanpu1 2
1. School of Computer Science and Engineering, Wuhan Institute of Technology, Wuhan 430205, China;2. Hubei Key Laboratory of Intelligent Robot(Wuhan Institute of Technology), Wuhan 430205,China;
关键词:
表情识别多任务学习知识图谱交叉压缩单元
Keywords:
facial expression recognition multi-task learning knowledge graph cross & compress units
分类号:
TN911.72
DOI:
10.19843/j.cnki.CN42-1779/TQ.202107010
文献标志码:
A
摘要:
针对面部表情分类的模型中参数较复杂、识别准确率较低的问题,提出了一种基于知识图谱辅助识别的多任务学习算法模型(MLAM),该模型由基于深度学习的识别模块与知识图谱嵌入模块两部分构成。首先从输入的数据中提取潜在的人脸局部表情特征,通过知识图谱实现局部表情和个体的复杂交互;然后在MLAM 模型中设计一个交叉压缩单元,关联这两个独立模块,自动学习局部表情和实体特征的高级交互,并在这两个任务之间传递交叉知识转移;最后,在FER2013和CK+的数据集上对比了同类算法,实验结果表明,该模型在上述数据集上分别得到了0.69和0.99的识别率,提高了面部表情识别准确率。
Abstract:
Aiming at the complicated parameters and low recognition accuracy rate in classification models of facial expression, this paper proposes a multi-task learning algorithm model (MLAM) based on knowledge graph assisted recognition,which consists of recognition module and knowledge graph embedding module based on deep learning. First, the potential facial expression features from the input data were extracted, and the complex interaction between the local expression and the individual through the knowledge graph was realized; then a cross & compress unit was designed in the MLAM, these two independent modules were associated, and the local expression and the advanced interaction of entity features were automatically learned and cross-knowledge between these two tasks was transferred. Finally, similar algorithms were compared on the FER2013 and CK+ data sets. The results showed that the model obtains recognition rates of 68.85% and 99.16% respectively on the above data sets, which improves the accuracy of facial expression recognition.

参考文献/References:

[1] 卢官明,何嘉利,闫静杰,等. 一种用于人脸表情识别的卷积神经网络[J]. 南京邮电大学学报(自然科学版),2016,36(1):16-22. [2] 吴晶晶,程武山. 融合全局和局部特征并基于神经网络的表情识别[J]. 计算机测量与控制,2018,26(6):172-175. [3] GIRSHICK R,DONAHUE J,DRRELL T,et al. Rich feature hierarchies for accurate object detection and demantic segmentation[C]// Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Seattle: IEEE Computer Society,2014: 580-587. [4] 尚骁,吴进. 采用Gabor-Houg变换的自适应滤波人眼定位[J]. 电讯技术,2016,56(3):324-330. [5] 胡建国. 基于局部Gabor二值化特征的实时表情识别系统[J]. 信息化研究,2015,41(3):68-71. [6] BISWAS S, SIL J, MAITY S P. PCA based face recognition on curvelet compressive measurements[C]//Proc of International Conference on Computation Intelligence, Communications, and Business Analytics. Singapore:Springer,2017:217-229. [7] KOCH?R ?M,GROSS M H, CARLS F R, et al. Simulating facial surgery using finite element? models[C]//Proceedings of the 23rd Annual Conference on Computer Graphics and Interactive Techniques.? New Orleans:ACM,1996:421-428. [8] 邹建成,邓豪. 一种基于卷积神经网络的人脸表情自动识别方法[J]. 北方工业大学学报,2019,31(5):51-56. [9] SCHROFF F, KALENICHENKO D, PHILBIN J.FaceNet:a unified embedding for face recognition and clustering[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Boston:IEEE,2015:815-823. [10] RIBEIRO E, UHL A, HAFNER M. Colonic polyp classification with convolutional neural networks[C]//Proceedings of 2016 IEEE 29th International Symposium on Computer-Based Medical Systems(CBMS). Dublin:IEEE:2016:253-258. [11] LI S,DENG W H,DU J P. Reliable crowdsourcing and deep local?ity-preserving learning for expression recognition in the wild[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu:IEEE, 2017:2584-2593. [12] 徐琳琳,张树美,赵俊莉. 构建并行卷积神经网络的表情识别算法[J]. 中国图象图形学报,2019(2):227-236. [13] 胡步发,王金伟. 双模态及语义知识的三维人脸表情识别方法[J]. 仪器仪表学报,2013,34(4):873-880. [14] 朱瑞,张俊三,朱杰,等. 一种基于KCNN和MKR的两阶段深度学习多任务推荐模型[J]. 陕西师范大学学报,2020(2):227-236. [15] KRIZHEVSKY A, SUTSKEVER I, HINTON G E, et al. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90. [16] WANG H, WANG N Y, YEUNG D Y. Collaborative deep learning for recommender systems[C]// In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2015:1235-1244. [17] WANG H W, ZHANG F Z, HOU M, et al. Signed heterogeneous information network embedding for sentiment link prediction[C]// In Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining. New York: ACM, 2018:592-600. [18] ZENG G,ZHOU J,JIA X, et al.Hand-crafted feature guided deep learning for facial expression recognition [C]// Proceedings of 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition.Xi’an:IEEE,2018:423-43.[19] HOWARD A, ZHMONGINOV A, CHEN L C, et al. Inverted residuals and linear bottlenecks: Mobile networks for classification, detection and segmentation[C]//The IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: Bostion: IEEE,2018:4510-4520. [20] WANG R X, FU B, FU G, et al. Deep & cross network for ad click predictions[C]//In Proceedings of the ADKDD’17. New York:ACM,2017:1-7.[21] KANADE T, TIAN Y, COHN F. Comprehensive database for facial expression analysis[C]//Proleedings Fourth of IEEE International Conference on Automatic Face & Gesture Recognition. Boston:IEEE,2000:46-53.

相似文献/References:

备注/Memo

备注/Memo:
收稿日期:2021-07-15基金项目:湖北省教育厅科学技术计划青年人才项目(Q20191514);武汉工程大学科学研究基金(16QD25,20QD32);湖北省大学生创新创业训练计划项目(S202110490040)作者简介:陈 龙,硕士研究生。E-mail:303549722@qq. com*通讯作者:张水平,博士,讲师。E-mail:zhangshuiping2007@126.com引文格式:陈龙,张水平,王海晖,等. 基于多任务学习和知识图谱的面部表情识别[J]. 武汉工程大学学报,2021,43(6):681-688.
更新日期/Last Update: 2021-12-27