Micro-expression recognition (MER) draws intensive research interest as micro-expressions (MEs) can infer genuine emotions. Prior information can guide the model to learn discriminative ME features effectively. However, most works focus on researching the general models with a stronger representation ability to adaptively aggregate ME movement information in a holistic way, which may ignore the prior information and properties of MEs. To solve this issue, driven by the prior information that the category of ME can be inferred by the relationship between the actions of facial different components, this work designs a novel model that can conform to this prior information and learn ME movement features in an interpretable way. Specifically, this paper proposes a Decomposition and Reconstruction-based Graph Representation Learning (DeRe-GRL) model to effectively learn high-level ME features. DeRe-GRL includes two modules: Action Decomposition Module (ADM) and Relation Reconstruction Module (RRM), where ADM learns action features of facial key components and RRM explores the relationship between these action features. Based on facial key components, ADM divides the geometric movement features extracted by the graph model-based backbone into several sub-features, and learns the map matrix to map these sub-features into multiple action features; then, RRM learns weights to weight all action features to build the relationship between action features. The experimental results demonstrate the effectiveness of the proposed modules, and the proposed method achieves competitive performance.
Octave Mix: Data augmentation using frequency decomposition for activity recognition
Graduate School of Engineering(研究生院);University of Fukui(福井大学)
在活动识别的研究领域,虽然很难收集大量的测量传感器数据,但关于数据增强(DA)的讨论还不多。在这项研究中,我提出了Octave Mix作为一种新的基于传感器的活动识别的综合风格的DA方法。Octave Mix是一种简单的DA方法,它通过使用频率分解将低频和高频波形交叉来组合两种类型的波形。此外,我提出了一个DA集成模型及其训练算法,以获得对原始传感器数据的鲁棒性,同时保持各种各样的特征表示。我进行了实验,以评估我提出的方法使用四个不同的基准数据集的基于感知的活动识别的有效性。因此,我提出的方法达到了最佳的估计精度。此外,我发现,集成两个DA策略:Octave Mix with rotation和mixup with rotation,可以实现更高的精度。
Orthogonal Deep Features Decomposition for Age-Invariant Face Recognition
Tencent AI Lab(腾讯人工智能实验室)
随着年龄的增长,人脸的外观会发生显著的类内变化,因此年龄不变人脸识别(AIFR)仍然是人脸识别领域的一个主要挑战。为了减少由年龄引起的类内差异,在本文中,我们提出了一种新的方法(即正交嵌入CNN,或OE-CNN)来学习年龄不变的深层人脸特征。具体来说,我们将深层人脸特征分解为两个正交分量来表示年龄相关和身份相关的特征。因此,对老化具有鲁棒性的身份相关特征随后用于AIFR。此外,为了补充现有的跨年龄人脸数据集,推进该领域的研究,我们构建了一个全新的大规模跨年龄人脸数据集(CAF)。在三个公共领域人脸老化数据集(MORPH Album 2、CACD-VS和FG-NET)上进行的大量实验表明了所提出方法的有效性和所构建的CAF数据集在AIFR上的价值。在最流行的通用人脸识别(GFR)数据集LFW上对我们的算法进行基准测试,还证明了GFR的泛化性能。
Decomposition-Based Domain Adaptation for Real-World Font Recognition
我们提出了一个域自适应框架,以解决合成训练和真实世界的测试数据之间的域不匹配。我们展示了我们的方法上一个具有挑战性的细粒度分类问题:识别字体风格的文本图像。在这个任务中,生成大量的渲染字体示例非常容易,但很难获得真实世界的标记图像。在以前的字体识别方法中,这种真实到合成域的差距导致了对新真实数据的泛化能力差(Chen et al.(2014))。在本文中,我们介绍了一种卷积神经网络分解方法,利用一个大型的训练语料库的合成数据,以获得有效的特征进行分类。这是使用基于堆叠卷积自动编码器的自适应技术来完成的,堆叠卷积自动编码器利用大量未标记的真实世界文本图像与以特定方式预处理的合成数据相结合。所提出的DeepFont方法在我们收集的一个新的大型标记真实世界数据集上实现了高于80%(前5名)的准确率。
University College London(伦敦大学学院);Chongqing Institute of Green and Intelligent Technology, CAS(中国科学院重庆绿色智能技术研究院);Southwest University(西南大学);Institute of Psychology, CAS(中国科学院心理研究所)
Micro-gesture Online Recognition using Learnable Query Points
School of Computer Science and Information Engineering(计算机与信息工程学院);School of Artificial Intelligence(人工智能学院);Hefei University of Technology (HFUT)(合肥工业大学);Key Laboratory of Knowledge Engineering with Big Data (HFUT), Ministry of Education(大数据知识工程教育部重点实验室);Institute of Artificial Intelligence, Hefei Comprehensive National Science Center(合肥综合性国家科学中心人工智能研究院);Zhejiang University(浙江大学);Anhui Zhonghuitong Technology Co., Ltd.(安徽中汇通科技有限公司)
ICE-GAN: Identity-aware and Capsule-Enhanced GAN with Graph-based Reasoning for Micro-Expression Recognition and Synthesis
University of Sydney(悉尼大学);University of New South Wales(新南威尔士大学)
微表情是人们真实情感和动机的反映,因此吸引了越来越多的研究者对人脸微表情自动识别的研究。短的检测窗口,微妙的面部肌肉运动和有限的训练样本使得微表情识别具有挑战性。为此,我们提出了一种新的基于图推理的身份感知和胶囊增强的生成对抗网络(ICE-GAN),引入微表情合成作为辅助任务来辅助识别。生成器生成具有可控微表情和身份感知特征的合成人脸,其长期依赖性通过图推理模块(GRM)捕获,并且该生成器检测图像真实性和表情类。我们的ICE-GAN在Micro-Expression Grand Challenge 2019(MEGC 2019)上进行了评估,与获奖者相比有显著改善(12.9%),并超过了其他最先进的方法。
DriverMHG: A Multi-Modal Dataset for Dynamic Recognition of Driver Micro Hand Gestures and a Real-Time Recognition Framework
TU Munich(慕尼黑工业大学);LMU Munich(慕尼黑大学);University of Tuebingen(图宾根大学);Intel Labs Europe(英特尔欧洲实验室);Intel Deutschland GmbH(英特尔德国有限公司)
手势的使用为人机交互(HCI)系统的笨重接口设备提供了自然的替代方案。然而,从视频流中实时的【形】动态微手势的识别对于车载场景是具有挑战性的,因为(i)手势应该自然地执行而不会分散驾驶员的注意力,(ii)微手势在空间受限区域处发生在非常短的时间间隔内,(iii)所执行的手势应该仅被识别一次,及(iv)整个架构应设计为轻量型,因为其将部署到嵌入式系统。在这项工作中,我们提出了一个HCI系统的动态识别驾驶员的微手势,这可能会产生至关重要的影响,特别是在汽车行业的安全相关的问题。为此,我们最初收集了一个名为Driver Micro Hand Gestures(DriverMHG)的数据集,该数据集由RGB、深度和红外模态组成。微手势的动态识别的挑战已经解决了提出一个轻量级的卷积神经网络(CNN)的基础架构,有效地运行在网上与滑动窗口的方法。对于CNN模型,应用了几种三维资源有效网络,并分析了它们的性能。手势的在线识别已与3D-MobileNetV 2,它提供了最好的离线准确性之间的应用网络具有类似的计算复杂性。最终的架构部署在驾驶员模拟器上,实时的【形】操作。我们公开了DriverMHG数据集和我们的源代码。
Feature refinement: An expression-specific feature learning and fusion method for micro-expression recognition
Jiangsu University(江苏大学);Nanjing Institute of Technology(南京工程学院);University of Oulu(奥卢大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);Xiamen University(厦门大学)
A New Multi-Domain Benchmark for Micro-Action Recognition and Detection
School of Computer Science and Information Engineering, Hefei University of Technology(合肥工业大学计算机科学与信息工程学院);School of Information Science and Technology, University of Science and Technology of China(中国科学技术大学信息科学技术学院)
Online Micro-gesture Recognition Using Data Augmentation and Spatial-Temporal Attention
School of Computer Science and Information Engineering, School of Artificial Intelligence, Hefei University of Technology (HFUT)(计算机科学与信息工程学院、人工智能学院,合肥工业大学);ReLER, CCAI, Zhejiang University, China(ReLER、CCAI、浙江大学,中国);Institute of Advanced Technology, University of Science and Technology of China(先进技术研究院,中国科学技术大学);Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, China(人工智能研究院,合肥综合性国家科学中心,中国);Key Laboratory of Knowledge Engineering with Big Data (HFUT), Ministry of Education(大数据知识工程重点实验室(HFUT),教育部);Xinsight Lab, Research Institute, Hefei Zhongjuyuan Intelligent Technology Co., Ltd., China(Xinsight实验室、研究院,合肥中 jouyuan智能科技有限公司,中国)
Benchmarking Micro-action Recognition: Dataset, Methods, and Applications
Hefei University of Technology (HFUT)(合肥工业大学);Hefei Comprehensive National Science Center, Institute of Artificial Intelligence(合肥综合性国家科学中心人工智能研究院);Anhui Zhonghuitong Technology Co., Ltd.(安徽中汇通科技有限公司);Lanzhou University(兰州大学)
Synergistic Spotting and Recognition of Micro-Expression via Temporal State Transition
School of Computer and Communication Engineering, University of Science and Technology Beijing(北京科技大学计算机与通信工程学院);School of Intelligence Science and Technology, University of Science and Technology Beijing(北京科技大学智能科学与技术学院)
Objective Class-based Micro-Expression Recognition through Simultaneous Action Unit Detection and Feature Aggregation
School of Computer Science and Communication Engineering, Jiangsu University(江苏大学计算机科学与通信工程学院);Department of Computer Science, Wayne State University(韦恩州立大学计算机科学系)
Three-Stream Temporal-Shift Attention Network Based on Self-Knowledge Distillation for Micro-Expression Recognition
MOEMIL Laboratory, School of Optoelectronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu, 611731, China(电子科学与技术大学光电科学与工程学院);School of Optoelectronic Science and Engineering, University of Electronic Science and Technology of China(电子科学与技术大学光电科学与工程学院);UESTC-MIT Joint Institute of Intelligent Microtechnique, Yibin, 644000, China(UESTC-MIT智能微技术联合研究所)
Facial Micro-Expression Spotting and Recognition using Time Contrasted Feature with Visual Memory
Kalyani Government Engineering College(卡利亚尼政府工程学院);Nanyang Technological University(南洋理工大学);Cognizant Technology Solutions(高知特科技解决方案公司);Indian Institute of Technology Roorkee(印度理工学院鲁尔基分校)
Motion Matters: Motion-guided Modulation Network for Skeleton-based Micro-Action Recognition
University College London(伦敦大学学院);ReLER, CCAI, Zhejiang University(ReLER、CCAI、浙江大学);Hefei University of Technology(合肥工业大学);Institute of Artificial Intelligence, Hefei Comprehensive National Science Center(人工智能研究院、合肥综合性国家科学中心);Intelligent Interconnected Systems Laboratory of Anhui Province(安徽省智能互联系统实验室)
Shallow Triple Stream Three-dimensional CNN (STSTNet) for Micro-expression Recognition
Feng Chia University(逢甲大学);National Taipei University of Nursing and Health Sciences(台北护理健康大学);Multimedia University(多媒体大学);Xinyang Normal University(信阳师范大学)