About me

I am currently an Associate Professor at Anhui University, Hefei, China, and also affiliated as a Researcher/PI with Tian-Hai Flow Control Co., LTD. Previously, I was a Postdoctoral Researcher at Yonsei University, Seoul, Korea, under the supervision of Prof. Andrew Teoh. My research there was funded by the South Korea Brain Pool (BP) Program, and my research at Monash was funded by the MARIE SKŁODOWSKA-CURIE Fellowship. I am proud to be a BP and MSCA fellowship awardee. My research expertise lies in image processing (both medical and natural), deep learning, computer vision, pattern recognition, and privacy protection.

I earned my Ph.D. degree from the Faculty of Information Technology, Monash University, and my B.Sc. degree with the National Training Base for Life Science and Technology Talents program at Huazhong Agricultural University. I was briefly enrolled as a Ph.D. student in molecular metabolism at Virginia Tech, USA. Since 2024, I have been a Editor/Guest Editor for several SCI journals and have served as an Area Chair (AC) for BMVC. Please feel free to use my Public PGP Key when you email me if you are interested in privacy protection.

I am proudly a participant in a charity project called 蓝信封行动 (Blue Letter Project), which supports children left behind in rural areas💖💖. I was interviewed about my involvement, and you can read the interview here.

Join Us!

👉 欢迎同学报考硕士研究生，申请博士研究生！一起探索 AI 的无限可能!博士综合待遇保底5000/月。金哲教授联合培养。所有硕博生课题组可视情况额外补助，顶级科研成果奖励另算。欢迎同学提前联系！（微信:xdongcs）

我们的四大核心研究方向 (Research Pillars)

mindmap
  root((我们))
    Security(🛡️ AI 安全与隐私)
      Cancelable_Biometrics(可撤销生物识别)
      Privacy_Preserving(隐私保护计算)
      Adversarial_Defense(对抗防御)
    Vision(👁️ AI 视觉增强)
      Low_Light(暗光增强<br/>Rawformer)
      Restoration(图像复原)
      Transformer(视觉 Transformer<br/>架构设计)
    Medical(🏥 AI 智慧医疗)
      Ultrasound(超声影像分析)
      Fetal_Heart(器官结构检测)
      Domain_Adaptation(域适应学习)
    Brain(🧠 AI 大脑与智能体)
      MLLM(多模态大模型)
      Agents(自主智能体)
      AIGC(生成式 AI)

1. AI 护盾：生物特征识别与安全 (Biometric Security & Privacy)

“在这个刷脸的时代，谁来保护你的‘脸’？”
核心：可撤销生物识别 (Cancelable Biometrics), 隐私保护人脸识别, 生物加密系统。
我们在做什么：设计能够抵御黑客攻击的“加密人脸”和“加密指纹”。即使数据库泄露，你的真实生物特征依然安全！这是信息安全领域的“防弹衣”。

2. AI 鹰眼：计算摄影与暗光增强 (Computational Photography & Enhancement)

“让黑夜如白昼，让模糊变清晰。”
核心：暗光增强 (Low-light Enhancement), 图像复原, Rawformer (基于Transformer的原始图像处理)。
我们在做什么：利用深度学习让手机在伸手不见五指的黑夜拍出清晰大片，修复老旧照片，提升监控画质。

3. AI 医者：医学影像分析 (Medical AI)

“AI 辅助诊断，守护生命健康。”
核心：超声图像分析 (Ultrasound Analysis), 胎儿心脏结构检测, 无监督域适应。
我们在做什么：教 AI 看懂复杂的 B 超图像，自动检测胎儿心脏异常，辅助医生进行早期诊断。做有温度的技术！

4. AI 大脑：大模型与智能体 (Large Models & Agents)

“赋予 AI 思考与自主决策的灵魂。”
核心：多模态大模型 (MLLM), 自主智能体 (Autonomous Agents), Generative AI.
我们在做什么：不仅仅是识别图像，我们让 AI 学会像人一样思考、规划和行动。将视觉 (Vision) 与语言 (Language) 结合，打造能够理解意图并自动执行任务的下一代智能体（Agent）。

💡 为什么选择我们？

可呈现的论文：我们在 CVPR, ICCV, TIFS等顶级会议和期刊上发表多篇高水平论文，访问谷歌学术查看详细论文列表。会手把手指导你发表第一篇顶会！绝对不放养。也欢迎和支持有独立能力成员开展独立研究，绝不瞎指导。
硬核：从底层算法架构设计到具体的医疗/安全应用，既有理论深度，又有应用广度。
有趣：与其做一个无情的调参机器，不如来做保护隐私的黑客克星，或者辅助医生的 AI 专家，或者智能体的构建者！

Dong Xingbo