Yunheng Liu
Papers
1
Total Citations
15
H-Index
1
About
Yunheng Liu is a rising researcher at the forefront of multimodal perception and quality assessment for 3D visual data. His work centers on the intersection of computer vision, large language models (LLMs), and graph-based learning, with a primary focus on point cloud quality assessment (PCQA)—a critical challenge for applications in autonomous driving, robotics, virtual reality, and 3D reconstruction. Liu’s most notable contribution is his pioneering 2024 paper, "LLM-Guided Cross-Modal Point Cloud Quality Assessment: A Graph Learning Approach," which has already garnered 15 citations. In this work, he introduces an innovative framework that leverages LLMs to guide cross-modal reasoning, using graph neural networks to model complex spatial and semantic relationships in degraded point clouds. This approach significantly advances the accuracy and reliability of PCQA, addressing a key bottleneck in real-world 3D systems. By bridging language understanding with geometric perception, Liu’s research opens new pathways for human-aligned quality metrics. His work is particularly impactful for students and engineers developing robust perception pipelines, demonstrating how LLMs can serve as powerful priors for 3D data analysis. Liu’s early-career achievements signal a promising trajectory in shaping next-generation multimodal AI systems.
Research Focus
Key Achievements
Top Papers
- 1