Lixing Xiao (肖力行)

Master Student

3D Vision Group
Zhejiang University
Hangzhou, China

Email: lixingxiao0 at gmail dot com

[Google Scholar] [GitHub] [Email]

Short Bio

I am a Master student at the 3D Vision Group, Zhejiang University, advised by Sida Peng and Xiaowei Zhou. Prior to joining ZJU, I earned my Bachelor of Engineering degree in Artificial Intelligence from Southeast University, where I built a solid foundation in computer vision, machine learning, and intelligent systems. My current research interests include Spatial Intelligence, Embodied AI and Agent systems.

News

[2026-04] One paper, NeAR, is selected as CVPR 2026 Highlight.
[2026-02] One paper accepted to CVPR 2026.
[2025-09] One paper accepted to NeurIPS 2025.
[2025-06] One paper, Go to Zero, is selected as ICCV 2025 Highlight.
[2025-06] Two papers accepted to ICCV 2025.

Selected Publications

MotionStreamer: Streaming Motion Generation via Diffusion-based Autoregressive Model in Causal Latent Space

Lixing Xiao, Shunlin Lu, Huaijin Pi, Ke Fan, Liang Pan, Yueer Zhou, Ziyong Feng, Xiaowei Zhou, Sida Peng^†, Jingbo Wang

ICCV 2025

[Paper] [Project Page] [Code]

Go to Zero: Towards Zero-shot Motion Generation with Million-scale Data

Ke Fan, Shunlin Lu, Minyue Dai, Runyi Yu, Lixing Xiao, Zhiyang Dou, Junting Dong, Lizhuang Ma, Jingbo Wang

ICCV 2025 (Highlight)

[Paper] [Project Page] [Code]

Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting

Nan Wang, Lixing Xiao, Yuantao Chen, Weiqing Xiao, Bohan Li, Zhaoxi Chen, Chongjie Ye, Shaocong Xu, Saining Zhang, Ziyang Yan, Pierre Merriaux, Lei Lei, Tianfan Xue, Hao Zhao^†

NeurIPS 2025

[Paper] [Project Page] [Code]

NeAR: Coupled Neural Asset–Renderer Stack

Hong Li, Chongjie Ye, Houyuan Chen, Weiqing Xiao, Ziyang Yan, Lixing Xiao, Zhaoxi Chen, Jianfeng Xiang, Shaocong Xu, Xuhui Liu, Yikai Wang, Baochang Zhang, Xiaoguang Han, Jiaolong Yang, Hao Zhao^†

CVPR 2026 (Highlight)

[Paper] [Project Page] [Code]

Open Source Project

SBTI-Buddy Skill

A Claude Code skill built on SBTI-Test — a personality framework designed specifically for programmers. While the original SBTI uses a questionnaire, SBTI Buddy skips the form entirely: it reads your AI conversation history and automatically determines your type. No questionnaire needed.

[GitHub], 2026

272-dim Motion Representation

This motion representation enables direct conversion from joint rotations to SMPL body parameters, removing the need for the error-prone inverse kinematics (IK) step.

[GitHub], 2025

Awards

[2023-2024] National Scholarship