A5100610474- Digital Media Forensic Detection
- Advanced Image Processing Techniques
- Image and Signal Denoising Methods
- Computer Graphics and Visualization Techniques
- Advanced Optical Imaging Technologies
- Advanced Vision and Imaging
- Open Education and E-Learning
- 3D Shape Modeling and Analysis
- High-Temperature Coating Behaviors
- Human Pose and Action Recognition
- Random lasers and scattering media
- Image Enhancement Techniques
- Smart Materials for Construction
- Infrastructure Maintenance and Monitoring
- Cryospheric studies and observations
- Optical Wireless Communication Technologies
- Climate change and permafrost
- Remote Sensing and LiDAR Applications
- Opportunistic and Delay-Tolerant Networks
- Anomaly Detection Techniques and Applications
- Phase Equilibria and Thermodynamics
- Vehicle License Plate Recognition
- Color Science and Applications
- Intelligent Tutoring Systems and Adaptive Learning
- Mobile Ad Hoc Networks
Beijing University of Posts and Telecommunications
2013-2024
Dalian University of Technology
2024
University of Shanghai for Science and Technology
2023
University of Science and Technology Beijing
2022
Xiamen University
2022
Space Engineering University
2021
Jilin University
2016
Tsinghua University
2004
Removing degradation from document images not only improves their visual quality and readability, but also enhances the performance of numerous automated analysis recognition tasks. However, existing regression-based methods optimized for pixel-level distortion reduction tend to suffer significant loss high-frequency information, leading distorted blurred text edges. To compensate this major deficiency, we propose DocDiff, first diffusion-based framework specifically designed diverse...
The goal of scene text image super-resolution is to reconstruct high-resolution text-line images from unrecognizable low-resolution inputs. existing methods relying on the optimization pixel-level loss tend yield edges that exhibit a notable degree blurring, thereby exerting substantial impact both readability and recognizability text. To address these issues, we propose TextDiff, first diffusion-based framework tailored for super-resolution. It contains two modules: Text Enhancement Module...
Autostereoscopic display technology, despite decades of development, has not achieved extensive application, primarily due to the daunting challenge three-dimensional (3D) content creation for non-specialists. The emergence Radiance Field as an innovative 3D representation markedly revolutionized domains reconstruction and generation, simplifying common users broadening applicability Light Displays (LFDs). However, combination these two technologies remains largely unexplored. standard...
Nucleation was the basis of fabrication two-dimensional materials in bottom-up deposition methods. However, classical nucleation theory (CNT) not suitable for supercritical fluid (SCF) because non-ideality SCF. Herein, a dilute solution system composed nonvolatile solute and carbon dioxide (scCO2) established pressure-induced phase (PI-SCPN) proposed. A new defined solute-solvent correlation function found to influence driving force lot this particular process, especially when...
Autostereoscopic display, despite decades of development, has not achieved extensive application, primarily due to the daunting challenge 3D content creation for non-specialists. The emergence Radiance Field as an innovative representation markedly revolutionized domains reconstruction and generation. This technology greatly simplifies common users, broadening applicability Light Displays (LFDs). However, combination these two fields remains largely unexplored. standard paradigm create...
Generative adversarial network (GAN) has become a hot research topic in the field of image processing. As an unsupervised training model, GAN been widely used computer vision, especially style transfer. The purpose is to make generator generate false image, and discriminator cannot tell whether input real or generated image. Compared with traditional models, model these advantages transfer: composed two different networks, loss function automatically learned by playing games each other....
The goal of scene text image super-resolution is to reconstruct high-resolution text-line images from unrecognizable low-resolution inputs. existing methods relying on the optimization pixel-level loss tend yield edges that exhibit a notable degree blurring, thereby exerting substantial impact both readability and recognizability text. To address these issues, we propose TextDiff, first diffusion-based framework tailored for super-resolution. It contains two modules: Text Enhancement Module...
In this paper, a scheme of laser-driven white light illumination based on diffuse reflection is proposed, which can effectively avoid the blue spot phenomenon. The with color temperature 4717 K, lumen 9.705 lm and rendering index 62 be obtained by using module 0.05 W laser excitation, consistent theoretical calculation results. We propose two methods to improve uniformity, namely hook face concentration gradient.