Jun Luo | 罗骏

Hello! I’m Jun Luo, an undergraduate student in Peking University, school of Physics. I’m currently taking a double major in computer science at school of EECS, PKU. I am an incoming Phd Student of Prof. Gang Zeng at PKU School of intelligence Science and Technology. I highly suggest reading this guide, which is guiding my research. Recently, I an interested in learning from and generative modeling of continuous data, like videos, images, and actions. My goal is to build intelligent machine that understand and interact with the world through language and action, thus imporving human's well being.
You can reach me by Email or Wechat. Welcome!

Email  /  Google Scholar  /  Github /  X /  WeChat /  CV

We introduce SceneAssistant, an agent that uses vision-language models and iterative visual feedback to generate and refine open-vocabulary 3D scenes from natural language descriptions.

We address zero shot domain adaptative semantic segmentation by leveraging diffusion model to generate two synthetic domains and using a progressive adaptation training strategy

I warmly welcome new friends. Besides discussing study, research, work, and other things that might make us anxious, feel free to chat with me about your hobbies. Sharing the same hobbies makes chatting more enjoyable. You can visit this page (written in my native language, Chinese) to learn more about me if you're interested.