About

Changwei Wang is a researcher at the forefront of embodied AI and robot perception, whose work bridges the critical gap between 3D scene understanding and multimodal intelligence. His most influential contribution, the MRFTrans framework (2024, 24 citations), introduces a novel Multimodal Representation Fusion Transformer that achieves state-of-the-art monocular 3D semantic scene completion—a foundational capability for autonomous navigation and manipulation. Complementing this, his comprehensive survey on multimodal fusion and vision-language models for robot vision (2025, 19 citations) has become an essential roadmap for researchers, synthesizing advances across perception, reasoning, and control. Wang’s technical depth extends to precise 6D pose estimation via the C2Fi-NeRF method (2024, 3 citations), which innovatively inverts neural radiance fields for coarse-to-fine object localization. Earlier in his career, he demonstrated systems-level thinking with the design of a non-planar coaxial rotor system for multi-rotor flying robots (2017, 3 citations), introducing face-to-face and back-to-back rotor configurations that improved aerodynamic efficiency. This trajectory—from aerial platform design to cutting-edge multimodal perception—positions Wang as a versatile researcher shaping how robots understand and interact with complex 3D environments through integrated vision-language reasoning.

Research Focus

Key Achievements

3
H-Index
5
Papers
56
Total Citations
11
Avg Citations/Paper
🏆 Most Cited Paper
MRFTrans: Multimodal Representation Fusion Transformer for monocular 3D semantic scene completion
24 citations · 2024
📈 Most Prolific Year: 2024 (2 Papers)
🤝 Key Collaborators: 24
🏛 Institutions: Chinese Academy of Sciences, Ministry of Public Security of the People's Republic of China, Qilu University of Technology, Fuzhou University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago