Michael Yuan | 袁承博

Michael (Chengbo) Yuan | 袁承博

🦊🦁🐮🐹🐰🐸🐶🐻🐨🐯

I am a Master student in Institute for Interdisciplinary Information Science (IIIS) at Tsinghua University, advised by Prof. Yang Gao🥳. Currently, my research interests mainly focus on Embodied AI (finding a way to achieve scalable robot learning), with the assistance of 3D computer vision (3DV).

Perviously, I have interned at MoonshotAI and MindSpore of Huawei. I am also interested in creating AI products or startups. I received my bachelor degree from the HongYi Honor class at the School of CS, Wuhan University😘, under the supervision of Prof. Yong Luo. I am also minoring in Law at School of Law and have a broad and extensive interest in philosophy and finance.

I welcome exploration, creation and excitement! 🤗🤩🤪

Google Scholar / Github / X(Twitter) / WeBlog / Zhihu

contact me: ycb24@mails.tsinghua.edu.cn michaelyuancb@163.com

Recent

We release RoboEngine, the first plug-and-play and generalizable robot segmentation and augmentation models !

A pathway I envision for scalable robot intelligence, which forms the central theme of my research.

Check out General Flow as a possible representation for scalable robot learning !

Research

My research interests primarily focus on Embodied AI and 3D Computer Vision, particularly in utilizing vision-based methods to develop intelligent robotic systems with scalable ways. My ultimate aspiration is to create general-propose robots😎 that can truly understand and interact with our physical world.
(representative papers are highlighted)

	RoboEngine: Plug-and-Play Robot Data Augmentation with Semantic Robot Segmentation and Background Generation Chengbo Yuan, Suraj Joshi, Shaoting Zhu, Hang Su, Hang Zhao, Yang Gao project page / arXiv / code / dataset In Submission, 2025* RoboEngine is the first plug-and-play visual robot data augmentation toolkit. Users can effortlessly generate physics-aware robot scenes with few lines code. This enable training only in one scenes and visual generalizing to almost arbitrary scenes.
	Self-Supervised Monocular 4D Scene Reconstruction for Egocentric Videos Chengbo Yuan, Geng Chen, Li Yi, Yang Gao project page / arXiv / code / thread In Submission, 2025 We train EgoMono4D, a fast, dense and generalizable 4D reconstruction model for egocentric videos with label-free self-supervised methods. The training is conducted sole on unlabeled videos, potentially to apply to more label-scarce fields in the future.
	General Flow as Foundation Affordance for Scalable Manipulation Learning Chengbo Yuan, Chuan Wen, Tong Zhang, Yang Gao project page / arXiv / code / openreview / thread Conference on Robot Learning (CoRL), 2024 We build a 3D flow prediction model directly from large-scale RGBD human video datasets. Based on this model, we achieve stable zero-shot human-to-robot skill transfer in the real world.
	Depression Diagnosis and Analysis via Multimodal Multi-order Factor Fusion Chengbo Yuan, Yongqian Li, Qiancheng Yang, Qianhui Xu, Yong Luo arXiv International Conference on Artificial Neural Networks (ICANN), 2024 A multi-order factor fusion framework is designed to fuse text, audio, and video information for depression diagnosis.
	MFT: Multi-scale Fusion Transformer for Infrared and Visible Image Fusion Chen-Ming Zhang, Chengbo Yuan, Yong Luo, Xin Zhou Paper International Conference on Artificial Neural Networks (ICANN), 2023 A pyramid network architecture for infrared and visible image fusion.

Project

	MindQuantFinance: High-Performance Quantitative Asset Pricing Library A library for asset pricing (Q-quant) based on MindSpore gitee This is an AI+Finance project I participated in, where we built a high-performance library for derivative pricing based on MindSpore. It offers features such as Black-Scholes calculations, Monte Carlo simulations, and solutions to Backward SDE.
	QFR: Tabular Prediction for Quantative Finance Research A tabular prediction library designed for financial stock prediction. github This is one of the outcomes of an interesting experience, during which I worked with Prof. Jian Li on stock prediction problems. It provide rich tools for factors extraction and results prediction.
	TaoClass: an AI-Driven Course Forum A course forum powered by Large Language Models (LLMs) and Vector Databases (VD). gitee This is a project I led, and it also serves as the final assignment for the Software Engineering course at Wuhan University. The forum utilizes AI technology to offer natural language course selection and intelligent back-end information management.