A5101647743- Robotics and Sensor-Based Localization
- Advanced Image and Video Retrieval Techniques
- Speech and Audio Processing
- Space Satellite Systems and Control
- Advanced Image Fusion Techniques
- Blind Source Separation Techniques
- Advanced Adaptive Filtering Techniques
- Remote-Sensing Image Classification
- Astro and Planetary Science
- Advanced Vision and Imaging
- Robotic Path Planning Algorithms
- Advanced Neural Network Applications
- Autonomous Vehicle Technology and Safety
- Remote Sensing and Land Use
- Planetary Science and Exploration
- Computer Graphics and Visualization Techniques
- Advanced Algorithms and Applications
- Traffic and Road Safety
- Traffic control and management
- Advanced Sensor and Control Systems
- Indoor and Outdoor Localization Technologies
- Traffic Prediction and Management Techniques
- Image and Signal Denoising Methods
- Vehicle License Plate Recognition
- Grey System Theory Applications
Wuhan Textile University
2025
Xidian University
2024
Southwest University
2024
Shanghai Changning Maternity and Infant Health Hospital
2024
East China Normal University
2024
Capital Normal University
2024
Nanjing University of Aeronautics and Astronautics
2017-2022
Northeastern University
2019-2021
University of California, Irvine
2004-2021
Nankai University
2021
Purpose To determine the status of depression and its key influencing factors among Chinese older adults in different living situations. Method Data 7,092 were obtained from China Health Retirement Longitudinal Survey. This study analyzed variables depressive symptoms using random forest modeling logistic regression. Results The prevalence is 33.2%, with highest alone (41.2%). Cognitive functioning life satisfaction are psychological that affect adults' mental health, whereas self-rated...
Recent advances in diffusion models have led to significant progress audio-driven lip synchronization. However, existing methods typically rely on constrained audio-visual alignment priors or multi-stage learning of intermediate representations force motion synthesis. This leads complex training pipelines and limited naturalness. In this paper, we present SayAnything, a conditional video framework that directly synthesizes movements from audio input while preserving speaker identity....
Intensive computation is entering data centers with multiple workloads of deep learning. To balance the compute efficiency, performance, and total cost ownership (TCO), use a field-programmable gate array (FPGA) reconfigurable logic provides an acceptable acceleration capacity compatible diverse computation-sensitive tasks in cloud. In this paper, we develop FPGA platform that leverages unified framework architecture for general-purpose convolutional neural network (CNN) inference at center....
In this paper we present a novel and efficient depth-image representation warping technique called DMesh which is based on piece-wise linear approximation of the as textured simplified triangle mesh. We describe application hierarchical multiresolution triangulation method to generate adaptively triangulated depth-meshes efficiently from reference depth-images, discuss depth-mesh segmentation methods avoid occlusion artifacts propose new hardware accelerated rendering that supports per-pixel...
Nowadays, AI has many applications in everyday human activities such as exercise, eating, sleeping, and automobile driving. Tech companies can apply to identify individual behaviors (e.g., walking, driving), analyze them, offer personalized feedback help individuals make improvements accordingly. While offering is more beneficial for drivers, most smart driver systems the current market do not use it. This paper presents AutoCoach, an intelligent agent that classifies drivers’ into different...
卷积神经网络(Convolutional Neural Network,CNN)和Transformer已被广泛应用于高分辨率遥感影像的建筑物提取任务。然而,CNN在建模长距离空间依赖时仍存在挑战,导致提取的建筑物存在内部空洞问题;而Transformer在捕捉空间局部细节特征上存在不足,容易导致建筑物边缘模糊及小型建筑物的漏检。为解决上述问题,本文提出了一种新型的双流网络模型用于高分辨率遥感影像的建筑物提取,名为ILGS-Net(Network for the Integration of Local and Global Features Stream)。该模型将CNN与Transformer相结合,采用多层级的局部-全局特征融合模块,实现了对建筑物的局部细节特征与全局上下文特征的高效融合。同时,在目标函数中引入边缘损失函数约束模型训练,提高了建筑物边界的定位精度。在三个高分辨率建筑物数据集上的实验结果显示,所提出方法的交并比均高于本文所对比的最佳方法,平均提高了1%。
A convex speech enhancement (CSE) method is presented based on optimization and pause detection of the sources. Channel spatial difference identified for enhancing each source individually while suppressing other interfering Sparse unmixing filters indicating channel differences are sought by <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">l</i> <sub xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> norm regularization split Bregman method....
In this research, a new fusion scheme based on improved regional energy contrast pyramid algorithm with respect to the human visual characteristics is presented, infrared and visible data from HJ1-B satellite, ASTER were performed in experiments. First of all, source are decomposed by transform. Second, energy, standard deviation similarity calculated, then operator was determined threshold deviation. Finally, fused image reconstructed inverse Moreover, based, Laplacian Pyramid Gradient...
Traditional Chinese medicine has usually been recognized to be efficacious treat chronic diseases from the western point-of-view. However, there is a long history in China of applying traditional many acute and urgent medical conditions. In this review, selected methods documented including blowing air ear, nose insufflating therapy, acupuncture moxibustion were presented as common practices promote consciousness recovery coma. We aimed explore mechanism these four with current scientific...