Jiajun Li

ORCID: 0009-0001-9227-9197
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Radiation Effects in Electronics
  • Geotechnical and Geomechanical Engineering
  • Medical Image Segmentation Techniques
  • Clay minerals and soil interactions
  • Aluminum toxicity and tolerance in plants and animals
  • Silicon Effects in Agriculture
  • Advanced Image and Video Retrieval Techniques
  • Image Retrieval and Classification Techniques
  • Pancreatic function and diabetes
  • Geomechanics and Mining Engineering
  • Dental Implant Techniques and Outcomes
  • AI and Big Data Applications
  • Integrated Circuits and Semiconductor Failure Analysis
  • Soil Geostatistics and Mapping
  • Brain Tumor Detection and Classification
  • Computational and Text Analysis Methods
  • AI in cancer detection
  • Bone Tissue Engineering Materials
  • Liver Disease Diagnosis and Treatment
  • Authorship Attribution and Profiling
  • Computer Graphics and Visualization Techniques
  • Endoplasmic Reticulum Stress and Disease
  • Dental materials and restorations
  • Image and Signal Denoising Methods
  • Advancements in Semiconductor Devices and Circuit Design

Liaoning Technical University
2025

Guangzhou University
2025

Chinese Academy of Sciences
2024

Paradigm Pharmaceuticals (United States)
2024

Xi'an Shiyou University
2024

Shanghai Cancer Institute
2024

Renji Hospital
2024

Shanghai Jiao Tong University
2024

Institute of Physics
2024

University of Chinese Academy of Sciences
2024

10.1109/cvpr52733.2024.00260 article EN 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2024-06-16

As the depth of coal mining in China continues to increase, fracturing rock masses has an increasingly complex impact on surrounding roadways. The majority mine's roadways run through with hard roofs and soft bottoms, which typically exhibit dynamic behaviour. To further research mechanical behaviour fracture evolution under hard-roof soft-floor conditions, study is based working faces a mine, have floors. Uniaxial compression tests were utilized properties circumstances, using acoustic...

10.1038/s41598-025-86839-6 article EN cc-by-nc-nd Scientific Reports 2025-01-29

We present RS-Diffusion, the first Diffusion Models-based method for single-frame Rolling Shutter (RS) correction. RS artifacts compromise visual quality of frames due to row-wise exposure CMOS sensors. Most previous methods have focused on multi-frame approaches, using temporal information from consecutive motion rectification. However, few approaches address more challenging but important single frame In this work, we an ``image-to-motion" framework via diffusion techniques, with a...

10.1609/aaai.v39i9.33015 article EN Proceedings of the AAAI Conference on Artificial Intelligence 2025-04-11

Text-to-image diffusion models have achieved tremendous success in the field of controllable image generation, while also coming along with issues privacy leakage and data copyrights. Membership inference arises these contexts as a potential auditing method for detecting unauthorized usage. While some efforts been made on models, they are not applicable to text-to-image due high computation overhead enhanced generalization capabilities. In this paper, we first identify conditional...

10.48550/arxiv.2405.14800 preprint EN arXiv (Cornell University) 2024-05-23

Computer assisted diagnostic technology has been widely used in clinical practice, specifically focusing on medical image segmentation. Its purpose is to segment targets with certain special meanings images and extract relevant features, providing reliable basis for subsequent diagnosis research. However, because of different shapes complex structures segmentation images, some imaging techniques have similar characteristics, such as intensity, color, or texture, organs tissues. The...

10.1615/critrevbiomedeng.2024052258 article EN Critical Reviews in Biomedical Engineering 2024-01-01

Image stitching from different captures often results in non-rectangular boundaries, which is considered unappealing. To solve current solutions involve cropping, discards image content, inpainting, can introduce unrelated or warping, distort non-linear features and artifacts. overcome these issues, we a novel diffusion-based learning framework, \textbf{RecDiffusion}, for rectangling. This framework combines Motion Diffusion Models (MDM) to generate motion fields, effectively transitioning...

10.48550/arxiv.2403.19164 preprint EN arXiv (Cornell University) 2024-03-28

Download This Paper Open PDF in Browser Add to My Library Share: Permalink Using these links will ensure access this page indefinitely Copy URL DOI

10.2139/ssrn.4826445 preprint EN 2024-01-01

<title>Abstract</title> Based on the structure of inverter chain, charge collection mechanism bipolar amplification effect NMOS device is studied. With reduction size, leads to failure original model due funnel and diffusion mechanism. damage single event transient (SET) MOS device, two classical reinforcement structures source extended drain are proposed, effectively reduces current electrode. The characteristics simulated, they compared discussed. It found that more obvious. Therefore,...

10.21203/rs.3.rs-5001077/v1 preprint EN cc-by Research Square (Research Square) 2024-10-10

Representing underwater 3D scenes is a valuable yet complex task, as attenuation and scattering effects during imaging significantly couple the information of objects water. This coupling presents significant challenge for existing methods in effectively representing both water medium simultaneously. To address this challenge, we propose Aquatic-GS, hybrid representation approach that represents medium. Specifically, construct Neural Water Field (NWF) to implicitly model parameters, while...

10.48550/arxiv.2411.00239 preprint EN arXiv (Cornell University) 2024-10-31
Coming Soon ...