About Me
Hi! I am XiangTai Yang, a second-year undergraduate student at Xi'an Jiaotong University, working on computer vision and machine learning.
My research interests lie in multimodal 3D vision and cross-modal learning, with a particular focus on aligning images and point clouds. More broadly, I am interested in developing models that generalize across diverse scenarios without relying on strong, task-specific inductive biases explicitly encoded in their design. I am particularly drawn to simple, unified, and end-to-end architectures, where structure and representations emerge from data rather than being manually imposed, enabling models to capture rich cross-modal relationships beyond what can be easily specified by human intuition.
Currently, I explore multimodal and cross-modal alignment for 3D vision, focusing on robust registration, reconstruction, and geometry-aware modeling, aiming to enable models to better perceive and reason about the 3D world.
Actively seeking graduate positions for Fall 2028. Feel free to reach out.