A5101512992- Image and Signal Denoising Methods
- Target Tracking and Data Fusion in Sensor Networks
- Geophysical Methods and Applications
- Fault Detection and Control Systems
- Seismic Imaging and Inversion Techniques
- Underwater Acoustics Research
- Simulation and Modeling Applications
- Inertial Sensor and Navigation
- COVID-19 diagnosis using AI
- Hydrological Forecasting Using AI
- Artificial Intelligence in Healthcare and Education
- Machine Learning and ELM
- Advanced Adaptive Filtering Techniques
- Advanced Control Systems Optimization
- Hand Gesture Recognition Systems
- Water Systems and Optimization
- Distributed Sensor Networks and Detection Algorithms
- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Radiomics and Machine Learning in Medical Imaging
- Multimodal Machine Learning Applications
- Control Systems and Identification
- Robot Manipulation and Learning
- Gaussian Processes and Bayesian Inference
- Domain Adaptation and Few-Shot Learning
Capital Normal University
2021-2025
Peking University
2024
Nanjing University of Science and Technology
2024
Beijing Institute of Technology
2020
Semi-supervised learning leverages insights from unlabeled data to enhance generalizability of the model, thereby decreasing dependence on extensive labeled datasets. Most existing methods overly focus local representations while neglecting global structures. On one hand, given that and images are presumed originate same distribution, it is probable similar regional features observed in both types correspond label. Current label propagation techniques, which predominantly propagate...
Neural Radiance Fields (NeRF) often struggle with reconstructing and rendering highly reflective scenes. Recent advancements have developed various reflection-aware appearance models to enhance NeRF's capability render specular reflections. However, the robust reconstruction of scenes is still hindered by inherent shape ambiguity on surfaces. Existing methods typically rely additional geometry priors regularize prediction, but this can lead oversmoothed in complex Observing critical role...
Neural Radiance Fields (NeRF) often struggle with reconstructing and rendering highly reflective scenes. Recent advancements have developed various reflection-aware appearance models to enhance NeRF's capability render specular reflections. However, the robust reconstruction of scenes is still hindered by inherent shape ambiguity on surfaces. Existing methods typically rely additional geometry priors regularize prediction, but this can lead oversmoothed in complex Observing critical role...
Abstract In the field of deep learning for medical image analysis, training models from scratch are often used and sometimes, transfer pretrained parameters on ImageNet is also adopted. However, there no universally accepted dataset specifically designed pretraining currently. The purpose this study to construct such a general validate its effectiveness downstream imaging tasks, including classification segmentation. work, we first build by collecting several public datasets (CPMID). And...
With the gradual development of virtual reality (VR) technology, users have increasingly higher demand for system interaction. The human-computer interaction (HMI) by means data glove is a widely used in current days. In this paper, type single-chip microcomputer based low cost designed figure posture acquisition. Also, programmable control hand model simulation. And they are building basic HMI system. terms hardware, adopts as core processing and collects, processes transmits together with...
In this paper, we shall discuss the convergence of continuous-discrete feedback particle filter (FPF) proposed in Yang et al. (2014 Yang, T., Blom, H. A., & Mehta, P. G. (2014). The time filter. 2014 American Control Conference (pp. 648–653). IEEE. [Google Scholar]). FPF is an interacting system N particles where interaction designed such that empirical distribution approximates posterior by innovation error-based control structure. Under some assumptions, it proved that, for a class...
In this paper, the filtering problem for general time-invariant nonlinear state-observation system is considered. Our work based on Yau-Yau framework developed by S.-T. Yau and third author in 2008. The key of how to compute solution forward Kolmogorov equation (FKE) off-line effectively. Motivated supervised learning machine learning, we develop an efficient method numerically solve FKE from point view optimization. Specifically, computation part, a reduced computing linear equations making...