Xiaohan Mao

Papers

Total Citations

H-Index

About

Xiaohan Mao is a rising star in the intersecting fields of computer vision, robotics, and embodied artificial intelligence. Her research centers on enabling AI agents to perceive, understand, and interact with 3D environments as humans do—through holistic, multi-modal perception. Her most notable contribution is the development of EmbodiedScan, a groundbreaking suite for multi-modal 3D perception that bridges the gap between raw sensor data and actionable language commands. This work, already cited over 50 times since its 2024 release, provides a comprehensive framework for embodied agents to explore spaces, recognize objects, and contextualize scenes from a first-person perspective. By integrating vision, language, and spatial reasoning, Mao’s research directly tackles the core challenge of making AI not just see, but truly understand and communicate about the physical world. Her work is pivotal for advancing robotics and human-AI interaction, positioning her as a key innovator in the push toward truly embodied intelligence.