Generating human motions: What it is, how it is commonly done, and how it should be done
Close your eyes and imagine yourself standing up from a chair, walking to a nearby bed, and sitting on the bed. Now open your eyes. I’m pretty sure that you exceled that task. However, it is not as simple a task for a computer to do it, even with today’s unbelievably strong generative AI technology. In this talk, I will share some recent progresses on human motion generation, and my personal understandings of how it should be modeled.
What does it feel like doing research in different envrionments
Research is a journey of exploration, discovery, and innovation, and the environments in which we conduct research can greatly impact our experiences and outcomes. One of the most exciting aspects of research is the chance to collaborate with people from different backgrounds and disciplines, each bringing their own perspective and approach to the work. By learning from one another, we can expand our knowledge and skills, and ultimately create more impactful and meaningful research. In this talk, I will explore what it feels like to do research in different environments and how we can harness the power of diversity to achieve our research goals.
Tengyu Liu is a researcher at Beijing Institute for General Artificial Intelligence. He obtained his PhD degree in computer science from UCLA in 2021 under the supervision of Prof. Song-Chun Zhu. Before that, he received his master’s degree in computer science from UCLA and bachelor’s degree in computer science from UIUC. His research interest lies in the intersection between 3D computer vision, computer graphics and robotics. His long term goal is to create intelligent agents that can interact with virtual or physical environments just like humans do. His recent works include generalizable dexterous grasping and dynamic and complex human object interactions.
Yixin Chen is a Research Scientist at Beijing Institute for General Artificial Intelligence (BIGAI). He received his M.S. and Ph.D. from University of California, Los Angeles (UCLA) advised by Professor Song-Chun Zhu. His research interests focus on computer vision and cognition, particularly in the areas of scene understanding, human-object interaction and social cue understanding.