Weiliang Meng

Papers

Total Citations

H-Index

About

Dr. Weiliang Meng is a leading researcher at the intersection of computer vision, multimodal learning, and robotic perception. His work centers on developing advanced fusion architectures that integrate diverse data modalities—such as images, point clouds, and language—to enable machines to understand and interact with complex 3D environments. Dr. Meng’s most notable contribution is the MRFTrans (Multimodal Representation Fusion Transformer), a pioneering framework for monocular 3D semantic scene completion that achieves state-of-the-art performance by synergizing visual and geometric cues. This work, published in 2024, has already garnered 24 citations, reflecting its immediate impact on autonomous driving and augmented reality. He also authored a comprehensive survey on multimodal fusion and vision-language models for robot vision (2025), which has accumulated 26 citations across two versions, establishing itself as a key reference for researchers seeking to bridge perception and reasoning in robotics. Dr. Meng’s research not only advances fundamental understanding of cross-modal representation learning but also provides practical blueprints for building more perceptive and context-aware robotic systems.