Chenming Zhu

Shanghai Artificial Intelligence Laboratory

Papers

Total Citations

H-Index

About

Chenming Zhu is a rising star at the intersection of computer vision, robotics, and embodied AI, whose work is redefining how machines perceive and interact with 3D environments. His landmark paper, "EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI" (2024), has already garnered 54 citations, signaling its rapid influence. Zhu’s core research focuses on enabling embodied agents—robots or AI systems—to explore, understand, and act within complex, real-world 3D spaces using first-person, multi-modal observations (like vision and language). His major contribution lies in bridging the gap between raw sensory data and high-level semantic understanding, creating holistic perception suites that allow agents to contextualize scenes for human interaction. This work is pivotal for advancing embodied AI, where agents must follow instructions and navigate dynamic environments. Zhu’s achievements include developing scalable frameworks that integrate 3D scanning, language grounding, and task-oriented reasoning, setting a new benchmark for the field. For students and researchers, his research offers a blueprint for building truly interactive, perceptive machines—a critical step toward autonomous robots and intelligent assistants.