I am an M.Sc. student at the Technical University of Munich and a research intern at Agile Robots SE. My research focuses on visual computing systems that model humans and 3D scenes over time—across reconstruction, controllable video generation, and perception-action world models for robots.
Yan Xia*†, Wenbo Ji*, Weirong Chen, Daniel Cremers
Overview
Matching-based fusion of sparse-view pointmaps into compact, cross-view-consistent 3D Gaussians.
LiteTracker: Leveraging Temporal Causality for Accurate Low-latency Tissue Tracking
MICCAI, 2025
Long-term Tracking
Authors
Mert Asim Karaoglu, Wenbo Ji, Ahmed Abbas, Nassir Navab, Benjamin Busam, Alexander Ladikos†
Overview
Causal temporal feature reuse with prior-motion initialization for accurate, low-latency online tissue tracking.
RE0: Recognize Everything with 3D Zero-shot Instance Segmentation
ICRA, 2025
Scene Decomposition
Authors
Xiaohan Yan*, Zijian Jiang*, Yinghao Shuai*, Nan Wang, Xiaowei Song, Wenbo Ji, Ge Wu, Jinyu He, Gang Wei, Zhicheng Wang†
Overview
Training-free 3D zero-shot instance segmentation from multi-view masks and CLIP semantics.
Thesis
Endoscopic Scene Reconstruction with 4D Half Gaussian Splatting
2025
Master's Thesis3D/4D Reconstruction
Overview
Developed a 4D Half-Gaussian splatting pipeline for deformable stereo endoscopic reconstruction with depth-prior initialization, HexPlane spatiotemporal deformation, and edge-aware depth regularization. Achieved 38.1 PSNR on EndoNeRF versus prior endoscopic GS/NeRF baselines; also evaluated on SCARED.
Experiences
Video World Model for Robot Dexterous ManipulationApr 2026 - Now
Research InternshipPerception-Action World Models
Contributions
Proposing a video-generation method that translates human egocentric manipulation videos into robot-domain videos to improve robot policy learning.
Mentors
Mahdi Mustapha HamadAgile Robots SE
Human Motion Video DiffusionMarch 2026 - Now
Master's ThesisVideo GenerationHuman Scene Interaction
Contributions
Developing camera-controlled human motion video diffusion for controllable synthesis of human-scene interactions.
Led method design and experiments for CSG-Fusion (matching-based pruning, mask-based feature aggregation, and cross-view supervision; ICCV Workshop E2E3D Best Paper).
On ScanNet++, +2.8 dB PSNR over Splatt3R at 90% overlap with ~124K fewer Gaussians; also evaluated zero-shot on DTU.
A static handbook for understanding how modern AI systems train, serve, generate, route, compress, and fail.
OpenUserStudyKit
March 2026
Personal Project
Focus
Reusable User Study Infrastructure
Overview
An open-source toolkit for building reusable user study questionnaires and experiment workflows.
LiteAvatar - WASM Version
Jan 2026
Personal Project
Focus
2D Audio-driven Human Avatar Animation
Overview
A lightweight audio-driven 2D avatar solution that runs entirely in the browser using WASM based on Lite-avatar. No backend server required.
Education
M.Sc. Electrical Engineering and Information Technology
Technical University of Munich2023 – Now
M.Sc. Electrical Engineering and Information Technology
Double-degree program with Tongji University (Tongji M.Sc. awarded 2025). Thesis: camera-controlled human motion video diffusion at the Visual Computing Group.
