Jun-Yan Zhu

2023 Fellow

Current Institution: Carnegie Mellon University

Computer/Information Sciences

Visit Lab Website

Connect with Jun-Yan Zhu:
Twitter LinkedIn

About Jun-Yan Zhu's Work

We are at a pivotal moment in our history. Large-scale AI models like DALL·E2 are transforming our society, making content creation accessible to all. However, this raises many pressing questions. Will AI replace human creators’ jobs? Is AI stealing artists’ work? How can AI and human creators work together? Who should get credited? My vision is to develop a human-centered generative modeling framework where human creators and generative models can co-exist and collaborate. In this framework, human creators can increase their creativity with the help of large-scale visual generative models while maintaining control over the creative process and receiving proper compensation. Our framework will create a paradigm shift in human-AI collaboration, benefiting not only computer vision, graphics, and medical imaging, but also other domains like conversational agents, assistive robotics, driver assistance systems, and healthcare.