A5101427731- Tensor decomposition and applications
- Visual Attention and Saliency Detection
- Advanced Neuroimaging Techniques and Applications
- 3D Shape Modeling and Analysis
- Advanced Surface Polishing Techniques
- RFID technology advancements
- Oil and Gas Production Techniques
- Advanced Sensor and Energy Harvesting Materials
- Conducting polymers and applications
- Fault Detection and Control Systems
- Structural Health Monitoring Techniques
- Advanced Neural Network Applications
- Industrial Vision Systems and Defect Detection
- Remote Sensing and Land Use
- Orthodontics and Dentofacial Orthopedics
- Geoscience and Mining Technology
- Sparse and Compressive Sensing Techniques
- Advanced MIMO Systems Optimization
- Engineering Diagnostics and Reliability
- Tactile and Sensory Interactions
- Manufacturing Process and Optimization
- Metal Extraction and Bioleaching
- Machine Learning and Algorithms
- Computer Graphics and Visualization Techniques
- Hydraulic Fracturing and Reservoir Analysis
Beijing Information Science & Technology University
2024
North China Electric Power University
2020
RIKEN Center for Advanced Intelligence Project
2019
Université Laval
2015-2016
Xi'an Jiaotong University
2011
Administration of Occupational Safety and Health
1984
Abstract Design of the capacitive tactile sensor with ultra-high sensitivity and fast response/recovery times is critical to advancement wearable devices. However, achieving both time simultaneously a huge challenge. In this work simple easy-to-prepare flexible presented, using biomimetic gray kangaroo structured dielectric layer polydimethylsiloxane. By finite element analysis study influences various structures, test result experimentally optimized showed (1.202 kPa −1 ), outstanding...
We present a novel generalized linear tensor regression model, which takes tensor-variate inputs as covariates and finds low-rank (almost) best approximation of coefficient arrays using hierarchical Tucker decomposition. With limited sample size, our model is highly compact extremely efficient it requires only O(dr <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> + dpr) parameters for order d tensors mode size p rank r, avoids the...
The higher-order partial least squares (HOPLS) is considered as the state-of-the-art tensor-variate regression modeling for predicting a tensor response from input. However, standard HOPLS can quickly become computationally prohibitive or merely impossible, especially when huge and time-evolving tensorial streams arrive over time in dynamic application environments. In this paper, we present efficient online algorithm, namely incremental (IHOPLS), adapting to setting of infinite...
Although the convolutional neural networks (CNNs) have become popular for various image processing and computer vision tasks recently, it remains a challenging problem to reduce storage cost of parameters resource-limited platforms. In previous studies, tensor decomposition (TD) has achieved promising compression performance by embedding kernel layer into low-rank subspace. However employment TD is naively on or its specified variants. Unlike conventional approaches, this paper shows that...
We propose a method for simulating cloth with meshes dynamically refined according to visual saliency. It is common belief that it preferable the regions of an image being viewed have more details than others. For certain scene, low-resolution mesh first simulated and rendered into images in preview stage. Pixel saliency values these are predicted pre-trained prediction model. These pixel saliencies then translated vertex corresponding meshes. Vertex saliency, together camera positions...
Abstract This research introduces an advanced algorithm based on convolutional neural networks for the detection and categorization of surface defects in manufacturing processes. At its core, employs a deep learning model that integrates residual attention mechanisms to effectively extract features. Additionally, we have developed novel feature selection method, named NR, which synergistically combines neighborhood component analysis ReliefF techniques. approach enables more representative...
A new method based on analyzing Elastic Supporter dynamic stress signals used to diagnose the rotor system faults in small and medium-sized aero-engine is proposed. Firstly, singular value decomposition (SVD) was de-noise elastic supporter signals, theory of difference energy entropy values achieved. This for determining proper number useful compared with sequence spectrum value. Secondly, de-noised were decomposed into a finite IMFs by EMD, order remove fictitious IMFs, spectral ratio...
This paper presents a novel method to compute vertex saliency from 3D facial meshes. Among the many descriptors for vertices, Spin-image correlation and curvature are used based on competition cooperation mechanisms. The proposed has been tested IAIR-3Dface Database evaluating generated salient regions Experimental results demonstrate that both abundance ratio accuracy of can be improved by mechanisms compared