
Ganlin Zhang
PhD StudentTechnical University of MunichSchool of Computation, Information and Technology
Informatics 9
Boltzmannstrasse 3
85748 Garching
Germany
Fax: +49-89-289-17757
Office: 02.09.044
Mail: ganlin.zhang@tum.de
Personal homepage: https://ganlinzhang.xyz
Research Interests
3D Vision, Visual SLAM, Structure from Motion, 3D Reconstrution.
Open Research Projects
Sequential 3D Reconstruction with 3D foundation model and Compact Scene Representation
Recent 3D foundation models (e.g., DUSt3R, VGGT) have demonstrated strong performance in reconstructing 3D scenes from RGB images. Follow-up works such as Spann3r and CUT3R have extended these approaches to sequential image data. However, most existing methods use per-frame point clouds as the scene representation, which leads to redundancy in overlapping regions.
This project aims to explore more compact scene representations, such as 3D Gaussian Splatting (3DGS), to reduce reconstruction redundancy and improve efficiency for sequential image data.
Preferred Requirements
- Strong research motivation, with an interest in producing publishable work.
- Solid background in 3D computer vision and multi-view geometry, preferably with related project experience.
- Proficiency in Python and PyTorch.
- Prior experience with SLAM/SfM, 3DGS, or 3D foundation models is a plus.
If you are interested, please contact me via email with your CV and academic transcript.
Publications
Export as PDF, XML, TEX or BIB
Preprints
2024
[] GlORIE-SLAM: Globally Optimized Rgb-only Implicit Encoding Point Cloud SLAM , In arXiv preprint arXiv:2403.19549, 2024. ([project],[code])
Conference and Workshop Papers
2025
[] Back on Track: Bundle Adjustment for Dynamic Scene Reconstruction , In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025. ([project page])
Best Paper Candidate [] Splat-slam: Globally optimized rgb-only slam with 3d gaussians , In Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR) Workshops, 2025. ([code])
2023
[] Revisiting Rotation Averaging: Uncertainties and Robust Losses , In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. ([code])