About Me

Hong Nie

Hong Nie

An undergraduate in Computer Science and Technology (the fifth round of major assessment: B+), Shanxi University ("Double Tops").

With the top comprehensive rank in my major, I am eligible for postgraduate recommendations in 2025 and have been admitted to the Department of Automation at the University of Science and Technology of China (USTC) for Artificial Intelligence (AI) master degree.

An acronym “gme” reminds me to be gorgeous, modest, and educated.

Interests: swimming, billiards, badminton, FPV, travel, and socializing.

Contact me: info@optichong.cn.

Research

My current research interests are MLLM(Multi-modal Large Language Model), Document AI, and Virtual Humans.

My undergraduate research focuses on Grasp Detection, Continual Learning, and Model Compression, guided by my undergraduate advisor, Chen Lu , an associate professor at the Institute of Big Data Science and Industry in Shanxi University.

Achievements related to scientific work are open-sourced on my Github.

Publication.

Academic

An enriching undergraduate experience.
  • A strong foundation in professional courses, qualifies me to be exempted from the postgraduate entrance examination. Rank: 1/117 (0.85%), CET-6 & CET-4: 540 & 587.
  • Participating in numerous computer-related competitions and emerging victorious.
  • Additionally, meaningful work to scientific research.
project report project report project report project report project report evening party to welcome newcomers develop project post-swimming

Portfolio

Postgraduate

Paper
Preprint
  1. Nie, H., Cao, F., Chen, L., Chen, F., Zou, Y., and Yu, J., “Few-Shot Identity Adaptation for 3D Talking Heads via Global Gaussian Field”, arXiv e-prints, Art. no. arXiv:2506.22044, 2025.
Print

Undergraduate

Paper
  1. Hong Nie, Zhou Zhao, Lu Chen*, et al. Smaller and Faster Robotic Grasp Detection Model via Knowledge Distillation and Unequal Feature Encoding. IEEE Robotics and Automation Letters, vol. 9, no. 8, pp. 7206-7213, Aug. 2024, doi: 10.1109/LRA.2024.3421790. (RA-L，IF：5.2，中科院二区TOP)
  2. L. Chen, Z. Li, Z. Lu, Y. Wang, H. Nie and C. Yang, “Domain-Invariant Feature Learning via Margin and Structure Priors for Robotic Grasping,” in IEEE Robotics and Automation Letters, vol. 10, no. 2, pp. 1313-1320, Feb. 2025, doi: 10.1109/LRA.2024.3520437.
Competition
  1. 第十四届中国大学生服务外包创新创业大赛        团队队长      国家级二等奖(赛题最高奖)
  2. 第十四届蓝桥杯大赛软件赛C++ B组      个人赛      国家级三等奖
  3. 2023年全国大学生数学建模竞赛       团队队长      省级一等奖
  4. 2022年全国大学生数学建模竞赛       团队队长      省级一等奖
  5. 第十七届中国大学生计算机设计大赛       团队副队长      省级二等奖
  6. 2022年华北五省(市、自治区)及港澳台大学生计算机应用大赛       团队队长      省级三等奖
Project

“共趣小斋” 智慧宿舍分配系统      项目负责人

基于2D虚拟人语音驱动算法的研究与应用      项目负责人

【Problem Statement】

A virtual digital human refers to an entity that exists in the non-physical world, created and utilized through computational means, and possesses multiple human-like characteristics (appearance features, human performance abilities, interaction capabilities, etc.). This problem primarily focuses on the 2D animation of real-world characters, specifically learning from videos of real individuals and using voice input to drive the character’s lip movements, facial expressions, and body actions.

A excellent solution can achieve the video of the character completing the new audio broadcast when the user provides the captured video and voice of the character speaking to obtain the new timbre and content of the audio. At the same time, the mouth motion of the person in this video should be based on being able to align the speech content.

【Additional Supplement】

In the output video driven by virtual human, the face of the character needs to be as clear as possible, and the eyes can blink naturally and randomly. At the same time, the teeth of the virtual human are not exactly the same as those in the original character video, but they should be clear and complete, and there should be no problems such as blurred teeth, multiple rows of teeth, and incomplete teeth.

【Reference】

Lip-Sync

Can’t see? Click me.

Can’t see? Click me.

Can’t see? Click me.

“AI Support” 养老服务系统      项目负责人

Patent
  • 软著      2D虚拟人语言驱动算法分析系统（2024SR0624977）      第一完成人
  • 软著      基于Android的养老服务软件（2024SR0624978）      第一完成人