Louis-Charles Caron

Institut national de recherche en sciences et technologies du numérique

Papers

Total Citations

H-Index

About

Louis-Charles Caron is a researcher whose work sits at the intersection of robotics, computer vision, and neural network architectures. His primary focus is on enabling robots to perceive and recognize objects in real-world environments by intelligently fusing multiple sensory data streams. Caron’s most significant contribution is a pioneering neural network-based fusion method that integrates color (2D) and depth (3D) information, along with spatial location, for robust object instance recognition on mobile robots. This approach, detailed in his 2015 paper, allows a robot to dynamically combine visual appearance with geometric cues, dramatically improving recognition accuracy in cluttered or poorly lit settings. The work has garnered 14 citations, reflecting its foundational role in the development of multi-modal perception systems. A key innovation is the coupling of this recognition architecture with a real-time 3D segmentation step, ensuring that the system can operate on live sensor feeds without pre-processing. Caron’s research directly addresses the challenge of bridging the gap between controlled laboratory conditions and the unstructured, dynamic environments where autonomous robots must operate, making his contributions highly relevant for advancing practical robotic applications.