Rongrong Ji
Papers
2
Total Citations
20
H-Index
2
About
Rongrong Ji is a leading figure in computer vision, with a career spanning foundational work in semantic segmentation to cutting-edge advances in 3D representation learning. His early research established efficient methods for image understanding, exemplified by his work on "Efficient semantic image segmentation with multi-class ranking prior" (2013, 12 citations), which tackled the critical challenge of parsing visual scenes with limited computational resources. More recently, Ji has been at the forefront of the rapidly evolving field of 3D representation, a domain pivotal for autonomous driving and robotics. His highly cited work "JM3D & JM3D-LLM: Elevating 3D Representation With Joint Multi-Modal Cues" (2024, 8 citations) directly confronts the limitations of simply transferring 2D learning strategies to 3D. By pioneering joint multi-modal cues, Ji is addressing fundamental challenges in how machines perceive and understand three-dimensional space, bridging the gap between 2D vision and the rich, volumetric data of the real world. His contributions are shaping the next generation of intelligent systems that must navigate and interact with complex physical environments.
Research Focus
Key Achievements
Top Papers
- 1Efficient semantic image segmentation with multi-class ranking prior12 citations · 2013
- 2