Zhihao Zhang (张志浩)

I am a first-year CS Ph.D. student focusing on 3D computer vision at Michigan State University (MSU), where my advisor is Prof. Xiaoming Liu. Before that, I graduated from Xi’an Jiaotong University (XJTU) with a Bachelor & Master degree in Computer Science, advised by Weizhan Zhang.

Prior to joining MSU, I interned at University of Illinois Urabana-Champaign (UIUC) with Prof. Yu-Xiong Wang.

Email  /  CV  /  Google Scholar  /  Twitter (X)  /  Github  / 

My research interests lie at the intersection of multi-modal learning and computer vision with the long-term goal of empowering computational models to better perceive and interact with the 3D visual world. Currently, I'm working on:

• 3D shape understanding, 3D object detection for autonomous driving
• Multi-modal learning for 3D perception

Highlights the understudied problem of Mono3D generalization to unseen ego heights.

Introduce TriAdapter Multi-Modal Learning (TAMM), a novel two-stage learning approach based on three synergistic adapters to different modalities in the pre-training.

Propose a tile classification based viewport prediction method with Multi-modal Fusion Transformer to improve the robustness of viewport prediction.