Vincent Cartillier
Papers
2
Total Citations
70
H-Index
2
About
Vincent Cartillier is a researcher at the forefront of embodied AI and semantic mapping, whose work bridges the gap between how machines perceive the world from a first-person view and how they build a coherent, bird’s-eye understanding of their surroundings. His most impactful contribution is the development of **Semantic MapNet**, a pioneering framework that enables an embodied agent—such as a robot or an egocentric AI assistant—to construct an allocentric, top-down semantic map of a new environment simply by taking a tour. This means the agent can answer “what is where?” by processing only egocentric RGB-D video and known pose data, a fundamental capability for autonomous navigation and scene understanding. With his seminal 2021 paper accumulating over 57 citations and an earlier version adding 13 more, Cartillier’s work has become a key reference in the field. His research directly addresses the challenge of spatial reasoning in AI, making it possible for machines to translate fleeting, personal observations into a stable, global representation of space—a critical step toward truly intelligent, context-aware assistants and robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2