A5107973961- Face and Expression Recognition
- Emotion and Mood Recognition
- Face recognition and analysis
- Biometric Identification and Security
- Remote-Sensing Image Classification
- Internet of Things and Social Network Interactions
- Advanced biosensing and bioanalysis techniques
- Technology and Security Systems
- Video Surveillance and Tracking Methods
- Advanced Biosensing Techniques and Applications
- Maritime Navigation and Safety
- UAV Applications and Optimization
- Image and Video Stabilization
- Technology and Data Analysis
- Robotic Path Planning Algorithms
- Gaze Tracking and Assistive Technology
- Advanced Chemical Sensor Technologies
- Nanoplatforms for cancer theranostics
- Anomaly Detection Techniques and Applications
Southwest University
2016-2024
Naval University of Engineering
2024
Institute of Psychology, Chinese Academy of Sciences
2019-2021
Chongqing Institute of Green and Intelligent Technology
2019
Civil Aviation Flight University of China
2012
Abstract In this paper, we investigate a deep learning vgg-16 network architecture for facial expression recognition under active near-infrared illumination condition and background. particular, consider the concept of transfer whereby features learned from high resolution images huge datasets can be used to train model relatively small dataset without loosing generalization ability. The pre-trained with technique has been trained validated on Oulu-CASIA NIR comprising six (6) distinct...
Owing to the vigorous development of face recognition, near-infrared (NIR) recognition technology with light insensitivity has attracted increasing attention. However, traditional methods for NIR feature hand-crafted design. In this paper, we present a convolutional neural network (CNN) recognition. CNN is multiplayer feed-forward which can automatically learn features from raw images and provide partial invariance illumination, scale deformation. Experimental results on PolyU-NIRFD database...
Near-infrared (NIR) facial expression recognition is resistant to illumination change. In this paper, we propose a three-stream three-dimensional convolution neural network with squeeze-and-excitation (SE) block for NIR recognition. We fed each stream different local regions, namely the eyes, nose, and mouth. By using an SE block, automatically allocated weights features further improve accuracy. The experimental results on Oulu-CASIA database showed that proposed method has higher rate than...
Facial expression recognition (FER) under active near-infrared (NIR) illumination has the advantages of invariance. In this paper, we propose a three-stream 3D convolutional neural network, named as NIRExpNet for NIR FER. The structure makes it possible to extract automatically, not just spatial features, but also, temporal features. design multiple streams enables fuse local and global facial To avoid over-fitting, moderate size suit Oulu-CASIA database that is medium-size database....
Face recognition based on depth images is widely studied due to its advantages of 3 dimensional information and environment illumination insensitivity. The traditional methods in this field mainly focus hand-crafted feature design, which cannot achieve satisfactory result. In addition, there no fixed face extraction method. To a better performance images, paper proposes method Convolutional Neural Networks(CNN). experiment performed database IIITD Kinect suggests that the proposed CNN...
Abstract The micro-expression spotting has recently attracted increasing attention from psychology and computer vision community, since embraced in the second facial Micro-Expression Grand Challenge (MEGC 2019). Different original feature difference (FD) analysis, this paper, we proposed a novel temporal spatial domain weight analysis of (TSW-FD) to achieve spotting. experimental results showed that TSW-FD improved 17.86% 24.21% F1-Score comparing FD CASME II SMIC-E-HS.
Facial expression, as a basic communication method, is an important way of emotion expression and cognition. emotional impairment seriously affects interpersonal social life. Micro-expressions (MEs) are involuntary instant facial dynamics that occurs when the subject failed to suppress their genuine emotions, especially in high-stake situations. Psychological research has shown MEs can reflect people's true which great help treatment impairment. ME spotting aims locate apex frame positions...