A5101749240- Color perception and design
- Sensor Technology and Measurement Systems
- Advanced Measurement and Metrology Techniques
- Optical measurement and interference techniques
- Sentiment Analysis and Opinion Mining
- Advanced MEMS and NEMS Technologies
- Advanced Fiber Optic Sensors
- Scientific Measurement and Uncertainty Evaluation
- Digital Media and Visual Art
- Image Enhancement Techniques
- Radar Systems and Signal Processing
- Optical Systems and Laser Technology
- Advanced SAR Imaging Techniques
- Advanced Bandit Algorithms Research
- Flow Measurement and Analysis
- AI and Multimedia in Education
- Machine Learning and Algorithms
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Extenics and Innovation Methods
- Survey Sampling and Estimation Techniques
Beihang University
2024
Hohai University
2020-2021
University of Virginia
2021
Gunma University
2010-2016
University of Science and Technology Beijing
2007-2012
Multimodal sentiment analysis (MSA) is an emerging research topic that aims to understand and recognize human or emotions through multiple modalities. However, in real-world dynamic scenarios, the distribution of target data always changing different from source used train model, which leads performance degradation. Common adaptation methods usually need data, could pose privacy issues storage overheads. Therefore, test-time (TTA) are introduced improve model at inference time. Existing TTA...
Multimodal sentiment analysis (MSA) is an emerging research topic that aims to understand and recognize human or emotions through multiple modalities. However, in real-world dynamic scenarios, the distribution of target data always changing different from source used train model, which leads performance degradation. Common adaptation methods usually need data, could pose privacy issues storage overheads. Therefore, test-time (TTA) are introduced improve model at inference time. Existing TTA...
Two-dimensional digital image correlation (2D-DIC) has the advantages of high computation efficiency and simple experimental setup compared to three-dimensional DIC (3D-DIC). However, 2D-DIC is sensitive out-of-plane motion, which leads rather poor strain results. Recently, optical extensometers realized by dual-reflector imaging have been proposed, can self-compensate errors introduced motion specimen. We will extend measurement from an extensometer full-field deformation measurement. To...
The present status and the future prospects of a method for precision mass force measurement, levitation (LMM), are reviewed. LMM has been proposed improved by authors. In LMM, inertial levitated using pneumatic linear bearing is used as reference applied to objects under test, such transducers, materials or structures. measured an optical interferometer. three typical applications i.e. dynamic calibration, micro material tester space scale, reviewed in this paper.
The present status and the future prospects of a method for precision mass force measurement, levitation (LMM), are reviewed. LMM has been proposed improved by authors. In LMM, inertial levitated using pneumatic linear bearing is used as reference applied to objects under test, such transducers, materials or structures. measured an optical interferometer. three typical applications i.e. dynamic calibration, micro material tester space scale, reviewed in this paper.
In heterogeneous rank aggregation problems, users often exhibit various accuracy levels when comparing pairs of items. Thus a uniform querying strategy over may not be optimal. To address this issue, we propose an elimination-based active sampling strategy, which estimates the ranking items via noisy pairwise comparisons from and improves users' average by maintaining set users. We prove that our algorithm can return true with high probability. also provide sample complexity bound for...