π This is Mingzhe Zheng, incoming 1st year PhD student @ HKUST, I'm current a researcher @ EverlynLab
π± My primary interests are computer vision, machine learning, and multimodality, especially in the direction of generative models (especially GAN & Diffusion). I aim to delve into efficient visual generation algorithms from a theoretical perspective and explore valuable applications of human-centric AIGC. My long-term research goal is to create virtual humans equipped with a trustworthy "brain" that can comprehend the surrounding environment and generate reactions (appearance-level, motion-level, speaking-level, etc.) to real-life interactions.
π€ I am learning LLMs currently and trying to use LLMs for visual-generative tasks in my recent paper and project.
π« I use Github as a repository for my coursework, projects and source codes for upcoming papers. By the way, I enjoy staring and following genius minds(mainly focus on splendid works in AIGC)
π Feel free to email me if there is any possibility of a conversation.
β‘ Keep learning and coding!