Runsen Xu

Papers

Total Citations

H-Index

About

Runsen Xu is a rising star in computer vision and robotics, whose work is central to advancing embodied AI—the quest to build intelligent agents that perceive, reason, and act in 3D environments. His research focuses on holistic multi-modal 3D perception, bridging the gap between raw sensor data and language-driven interaction. Xu’s most notable contribution is **EmbodiedScan**, a pioneering suite introduced in 2024 that provides a comprehensive framework for embodied agents to understand 3D scenes from first-person observations. This work, already garnering 54 citations, addresses a critical bottleneck: enabling machines to integrate visual, spatial, and linguistic cues for real-world tasks. By creating a unified benchmark and dataset, Xu has set a new standard for evaluating how well AI systems can “see” and “talk” about their surroundings. His research is instrumental in moving beyond static image recognition toward dynamic, interactive understanding—a key step for applications in robotics, augmented reality, and autonomous navigation. With his innovative approach to multi-modal learning, Runsen Xu is shaping the future of how machines perceive and interact with the physical world.