Haibin Huang

I recently joined TeleAI as a Principal Research Scientist and Director of World Model Research Center, leading research in multi-modal generative AI. Previously, I served as a Generative AI Researcher at ByteDance, Staff Research Scientist at Kuaishou Technology, and Research Scientist at Megvii Research.

I hold a Ph.D. in Computer Science from the University of Massachusetts Amherst, where I worked with Prof. Evangelos Kalogerakis and Prof. Rui Wang in the Computer Graphics Research Group. I earned my bachelor's and master's degrees from the Department of Mathematics at Zhejiang University.

My research focuses on the analysis and creation of visual content through the integration of computer vision, computer graphics, and machine learning. I aim to develop unified generative frameworks that bridge perception and creation, enabling controllable and high-quality content generation. I have a strong track record of translating research into real-world impact across TeleAI, ByteDance, Kuaishou, and Megvii, delivering scalable AI tools used by millions. Ultimately, my goal is to democratize digital creation—making it intuitive and accessible, and lowering the barrier between imagination and expression.

News

Jun 2026 Released SwiftVR, a real-time one-step generative video restoration model for streaming-quality video enhancement.
May 2026 Released Full-4D, a single-view video-to-4D framework for generating full-scope 4D scenes.
May 2026 Released TelePhysics, a physics-grounded multi-object scene generation framework from a single image with real-time interaction.
Feb 2026 4 paper accepted to CVPR 2026.

Our team is hiring full-time research scientists and interns specializing in generative AI (Shanghai/Beijing), with a focus on Multi-Modal generation and world models. Drop me an email if you are interested!

July 2025 - Present
Principal Research Scientist & Director of World Model Research Center

Institute of Artificial Intelligence, China Telecom (TeleAI)
March 2024 - July 2025
Researcher of Generative AI

ByteDance
Nov 2019 - March 2024
Senior Staff Research Scientist

Kuaishou Technology
Oct 2017 - Nov 2019
Research Scientist

Megvii Research

View Full Publication List →

SwiftVR: Real-Time One-Step Generative Video Restoration

Jiaqi Yan, Xiangyu Chen, Xinlin Zhong, Haibin Huang, Chi Zhang, Jie Liu, Jiantao Zhou, Xuelong Li

Project Page / Paper / Code
Full-4D: Generating Full-Scope 4D Scenes from a Single-View Video

Tingxi Chen, Ke Hao, Yabo Chen, Zhengxue Cheng, Rong Xie, Li Song, Haibin Huang, Chi Zhang, Xuelong Li

Project Page / Paper
TelePhysics: Physics-Grounded Multi-Object Scene Generation from a Single Image with Real-Time Interaction

Xin Zhang, Yabo Chen, Yijie Fang, Wanying Qu, Haibin Huang, Chi Zhang, Feng Xu, Xuelong Li

Project Page / Paper
TeleBoost: A systematic alignment framework for high-fidelity, controllable, and robust video generation

TeleBoost Team

Project Page
TeleStyle: Content-Preserving Style Transfer in Images and Videos

Shiwen Zhang, Xiaoyan Yang, Bojia Zi, Haibin Huang, Chi Zhang, Xuelong Li

Project Page
TeleWorld: Towards Dynamic Multimodal Synthesis with a 4D World Model

TeleWorld Team

ArXiv Preprint 2026
Macro-from-Micro Planning for High-Quality and Parallelized Autoregressive Long Video Generation

Xunzhi Xiang, Yabo Chen, Guiyu Zhang, Zhongyu Wang, Zhe Gao, Quanming Xiang, Gonghu Shang, Junqi Liu, Haibin Huang, Yang Gao, Chi Zhang, Qi Fan, Xuelong Li

Project Page
ATI: Any Trajectory Instruction for Controllable Video Generation

Angtian Wang, Haibin Huang, Jacob Zhiyuan Fang, Yiding Yang, Chongyang Ma

Project Page

Editorial Services

Associate Editor: The Visual Computer (2022-Present)

Program Committee

SIGGRAPH Technical Papers Committee 2025
SIGGRAPH Asia Technical Communications and Posters 2024, 2025
Eurographics Short Papers 2024, 2025
Computational Visual Media (CVM) 2023 - 2026
Shape Modeling International (SMI) 2020 - 2025