A5101924786- Face recognition and analysis
- Face and Expression Recognition
- Chaos-based Image/Signal Encryption
- Emotion and Mood Recognition
- Advanced Steganography and Watermarking Techniques
- Cleft Lip and Palate Research
- Mental Health via Writing
- Biometric Identification and Security
- 3D Shape Modeling and Analysis
- Image Processing Techniques and Applications
- Advanced Data Compression Techniques
- Advanced Numerical Analysis Techniques
- Computer Graphics and Visualization Techniques
- Video Surveillance and Tracking Methods
- Advanced Image and Video Retrieval Techniques
- CCD and CMOS Imaging Sensors
- Image and Signal Denoising Methods
- Medical Image Segmentation Techniques
- Advanced Vision and Imaging
- Image Retrieval and Classification Techniques
- Visual Attention and Saliency Detection
- Cryptographic Implementations and Security
- Image and Object Detection Techniques
- Fractal and DNA sequence analysis
- Gaze Tracking and Assistive Technology
Ludong University
2025
Capital Normal University
2015-2024
Beijing Advanced Sciences and Innovation Center
2016-2023
National Astronomical Observatories
2008-2014
Polytechnic University of Turin
2014
Chinese Academy of Sciences
2014
As a severe psychiatric disorder disease, depression is state of low mood and aversion to activity, which prevents person from functioning normally in both work daily lives. The study on automated mental health assessment has been given increasing attentions recent years. In this paper, we the problem automatic diagnosis depression. A new approach predict Beck Depression Inventory II (BDI-II) values video data proposed based deep networks. framework designed two stream manner, aiming at...
Recent evidence in mental health assessment have demonstrated that facial appearance could be highly indicative of depressive disorder. While previous methods based on the analysis promise to advance clinical diagnosis disorder a more efficient and objective manner, challenges visual representation complex depression pattern prevent widespread practice automated diagnosis. In this paper, we present deep regression network termed DepressNet learn with explanation. Specifically, convolutional...
In this paper, we present an automatic kinship verification system based on facial image analysis under uncontrolled conditions. While a large number of studies human face have been performed in the literature, there are few attempts for verification, possibly due to lacking such publicly available databases and great challenges problem. To end, collect database by searching 400+ pairs public figures celebrities from internet, automatically detect them with Viola-Jones detector. Then,...
This paper presents a Gabor-based Gradient Orientation Pyramid (GGOP) feature representation method for kinship verification from facial images. First, we perform Gabor wavelet on each face image to obtain set of magnitude (GM) images different scales and orientations. Then, extract the (GOP) GM multiple fusion verification. When combined with discriminative support vector machine (SVM) classifier, GGOP demonstrates best performance in our experiments, comparison several state-of-the-art...
Kinship verification from facial images is a challenging problem in computer vision, and there very few attempts on tackling this the literature. In paper, we propose new neighborhood repulsed metric learning (NRML) method for kinship verification. Motivated by fact that interclass samples (without relations) with higher similarity usually lie are more easily misclassified than those lower similarity, aim to learn distance under which intraclass (with pushed as close possible lying pulled...
Automatic depression diagnosis is a challenging problem, that requires integrating spatial-temporal information and extracting features from audio-visual signals. In terms of privacy protection, the development trend recognition algorithms based on facial landmarks has created additional challenges difficulties. this paper, we propose an attention network (AVA-DepressNet) for recognition. It novel multimodal framework with uses attention-based modules to enhance spatial temporal features....
Depression is a serious mental disorder that has received increased attention from society. Due to the advantage of easy acquisition speech, researchers have tried propose various automatic depression recognition algorithms based on speech. Feature selection and algorithm design are main difficulties in speech-based recognition. In our work, we spatial–temporal feature network (STFN) for recognition, which can capture long-term temporal dependence audio sequences. First, obtain better...
Recent visual-based depression recognition methods mostly use hand-crafted features with information lost in color channels, or deep network a limited performance from the finite data. In this paper, we propose method called Local Quaternion and Global Deep Network (LQGDNet) which can combine advantages features. Specifically, XOR Asymmetrical Regional Gradient Coding (XOR-AR-LGC) is first designed, encodes facial images local textures quaternion domain to keep dependence of integrated into...
Kinship verification from facial images in wild conditions is a relatively new and challenging problem face analysis. Several datasets algorithms have been proposed recent years. However, most existing are of small sizes one standard evaluation protocol still lack so that it difficult to compare the performance different kinship methods. In this paper, we present Verification Wild Competition: first competition which held conjunction with International Joint Conference on Biometrics 2014,...
Kinship verification from face images is a new and challenging problem in pattern recognition computer vision, it has many potential real-world applications including social media analysis children adoptions. Most existing methods for kinship assume that each positive pair of (with kin relationship) greater similarity score than those negative pairs without relationships under distance metric to be learned. In practice, however, this assumption usually too strict real-life samples. Instead,...
Related studies have revealed that the phonological features of depressed patients are different from those healthy individuals. With increasing prevalence depression, objective and convenient early screening is necessary. To this end, we propose an automatic depression detection method based on hybrid speech extracted by deep learning, dubbed as TTFNet. Firstly, to effectively excavate intrinsic relationship among multidimensional dynamic in frequency domain, Mel spectrogram raw its related...
Kinship verification via facial images is an emerging problem in computer vision and biometrics. Recent research has shown that learning a kin similarity measurement plays critical role constructing vision-based kinship system. We propose this paper new metric method for on human faces. To end, we first extract multiple feature representations each face image using different descriptors. Then, sparse bilinear models (one view) are jointly learned by joint structured sparsity-inducing norms,...
Motivated by the key observation that children generally resemble their parents more than other persons with respect to facial appearance, distance metric (similarity) learning has been dominant choice for state-of-the-art kinship verification via images in wild. Most existing learning-based approaches verification, however, are focused on a genetic similarity measure batch manner, leading less scalability practical applications ever-growing amount of data. To address this, we propose new...