Zisen Shao

Zisen Shao 邵子森

I am a first-year master's student in Computer Science at the University of Maryland, College Park, where I am advised by Prof. Ruohan Gao.

Previously, I earned my bachelor's degree in Computer Science from the University of Wisconsin–Madison, where I received advice from Prof. Josiah Hanna. I also participated in WisTex United at RoboCup, led by Prof. Josiah Hanna and Prof. Peter Stone.

I'm interested in computer vision and computer graphics. My research focuses on physics-informed models that can (1) reconstruct the digital twins of our Multimodal World from multimodal data, and (2) leverage shared representations across modalities to enable cross-modal learning.

Email / Github / LinkedIn

Aurora over Lake Mendota

Publications

	Objects as Audio-Visual Modal Sound Fields Zisen Shao^, Zihao Wei^, Derong Jin, and Ruohan Gao European Conference on Computer Vision (ECCV), 2026 project page / arXiv / code A novel object-level acoustic representation reconstructed from multi-view images and only a few impact sound recordings.
	FreeFix: Boosting 3D Gaussian Splatting via Fine-Tuning-Free Diffusion Models Hongyu Zhou, Zisen Shao, Sheng Miao, Pan Wang, Dongfeng Bai, Bingbing Liu, Yiyi Liao International Conference on 3D Vision (3DV), 2026 project page / arXiv / code A fine-tuning-free approach designed to eliminate artifacts and boost the rendering quality of 3D Gaussian Splatting (3DGS) in extrapolated views
	WeRef: An Open-source and Extensible Dataset for Referee Gesture Recognition in RoboCup Zisen Shao, Josiah P. Hanna, RoboCup-2025: Robot Soccer World Cup XXVIII, 2025 (Oral Presentation) code WeRef is an open-source synthetic data generation pipeline for RoboCup Standard Platform League (SPL) referee gestures recognition.
	Reinforcement Learning Within the Classical Robotics Stack: A Case Study in Robot Soccer Adam Labiosa, Zhihan Wang, Siddhant Agarwal, William Cong, Geethika Hemkumar, Abhinav Narayan Harish, Benjamin Hong, Josh Kelle, Chen Li, Yuhao Li, Zisen Shao, Peter Stone, Josiah P. Hanna IEEE International Conference on Robotics and Automation (ICRA), 2025 (Champion of RoboCup SPL Challenge Shield Division 2024, Best RoboCup-Themed Paper Award @ Roboletics 2.0) arXiv / code / media A novel architecture integrating RL within a classical robotics stack, while employing a multi-fidelity sim2real approach and decomposing behavior into learned sub-behaviors with heuristic selection.

Thanks to this amazing template!