Xiaoshuai Sun

Papers

Total Citations

H-Index

About

Xiaoshuai Sun is a leading researcher in computer vision, with a primary focus on 3D representation learning, multi-modal perception, and 3D object reconstruction. His work addresses critical challenges in how machines understand and reconstruct three-dimensional environments, a field essential for advancements in autonomous driving, robotics, and augmented reality. Sun’s major contributions include pioneering the integration of joint multi-modal cues—combining visual, textual, and geometric data—to elevate 3D representations beyond traditional 2D alignment strategies. His highly cited 2024 paper, "JM3D & JM3D-LLM," tackles the limitations of straightforward 2D-to-3D transfer, proposing novel frameworks that enhance generalization and information richness. This work has already garnered 8 citations, reflecting its immediate impact on the community. Earlier, Sun explored 3D object reconstruction from stereo images, identifying and addressing the poor generalization of methods that rely on memorizing training models. His research consistently pushes toward more robust, scalable 3D understanding, making him a notable figure in the evolution of next-generation computer vision systems.