Postdoctoral Researcher at Technical University of Munich (TUM) | Chief Technology Officer at VisualAIs
I work on 3D computer vision, with a focus on scene reconstruction, 3D scene understanding, and semantic scene graphs. My current postdoctoral research at TUM's CAMP Chair investigates 3D foundation models for scene reconstruction. At VisualAIs, I develop methods for realistic 3D rendering from smartphone captures.
I completed my PhD (Dr. rer. nat.) at TUM in March 2026. Previously, I was a research intern at NVIDIA Research (Taipei), working on prompt-based image editing using large language models. My background also includes research software engineering, SLAM, augmented reality, and technology entrepreneurship.
- 3D scene reconstruction and understanding
- 3D foundation models and neural rendering
- Semantic scene graphs and geometric learning
- Visual localization and SLAM
- Generative AI for visual content
| Project | Description | Publication |
|---|---|---|
| 3DSSG | Incremental 3D semantic scene graph prediction from RGB sequences | CVPR 2023 |
| SceneGraphFusion | Incremental 3D scene graph prediction from RGB-D sequences | CVPR 2021 |
| SCFusion | Real-time incremental scene reconstruction with semantic completion | 3DV 2020 |
- VoxNeRF: Bridging Voxel Representation and Neural Radiance Fields for Enhanced Indoor View Synthesis. IEEE Robotics and Automation Letters, 2025. DOI
- SCRREAM: Scan, Register, Render and Map: A Framework for Annotating Accurate and Dense 3D Indoor Scenes with a Benchmark. NeurIPS, 2024. Code
- CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graph Diffusion. NeurIPS, 2024. Code
- Incremental 3D Semantic Scene Graph Prediction from RGB Sequences. CVPR, 2023. Code
- SceneGraphFusion: Incremental 3D Scene Graph Prediction from RGB-D Sequences. CVPR, 2021. Code