Xiaoni Cai
Papers
1
Total Citations
24
H-Index
1
About
Xiaoni Cai is a rising star in embodied AI and robotic manipulation, whose work centers on enabling robots to intelligently interact with and rearrange objects in complex environments. Her most notable contribution is the development of SG-Bot, a groundbreaking framework for object rearrangement that employs a coarse-to-fine robotic imagination approach built on scene graphs. This innovative method allows robots to understand spatial relationships and plan manipulation tasks with unprecedented efficiency, moving beyond simple pick-and-place to context-aware rearrangement. The paper, published in 2024, has already garnered 24 citations, signaling its rapid impact on the field. Cai's research bridges the gap between scene understanding and physical interaction, addressing a core challenge in embodied AI: how robots can reason about object semantics and geometry simultaneously. Her work is particularly significant for applications in domestic robotics and automated warehousing, where adaptable object rearrangement is critical. By integrating scene graph representations with robotic control, Cai is helping to define a new paradigm for how machines perceive and modify their physical surroundings, making her a researcher to watch in the evolving landscape of intelligent robotics.
Research Focus
Key Achievements
Top Papers
- 1