|
3DGS-LM: Faster Gaussian-Splatting Optimization with Levenberg-Marquardt
Lukas Höllein, Aljaž Božič, Michael Zollhöfer, Matthias Nießner,
ICCV, 2025
project page
/
arXiv
/
video
/
code
3DGS-LM accelerates Gaussian-Splatting optimization by replacing the ADAM optimizer with Levenberg-Marquardt.
|
|
QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization
Yueh-Cheng Liu, Lukas Höllein, Matthias Nießner, Angela Dai
ICCV, 2025
project page
/
arXiv
/
video
QuickSplat learns data-driven priors to generate dense initializations for 2D gaussian splatting optimization of large-scale indoor scenes. We further learn to jointly estimate the densification and update of the scene parameters during each iteration.
|
|
ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models
Lukas Höllein, Aljaž Božič, Norman Müller, David Novotny, Hung-Yu Tseng, Christian Richardt, Michael Zollhöfer, Matthias Nießner
CVPR, 2024
project page
/
arXiv
/
video
/
code
ViewDiff generates high-quality, multi-view consistent images of a real-world 3D object in authentic surroundings.
|
|
ControlRoom3D: Room Generation using Semantic Proxy Rooms
Jonas Schult, Sam Tsai, Lukas Höllein, Bichen Wu, ..., Ji Hou
CVPR, 2024
project page
/
arXiv
/
video
ControlRoom3D creates diverse and plausible 3D room meshes aligning well with user-defined room layouts and textual descriptions of the room style.
|
|
Text2Room: Extracting Textured 3D Meshes from 2D Text-to-Image Models
Lukas Höllein*, Ang Cao*, Andrew Owens, Justin Johnson, Matthias Nießner
ICCV, 2023 (Oral Presentation)
project page
/
arXiv
/
video
/
code
Text2Room generates textured 3D meshes from a given text prompt using 2D text-to-image models. The core idea of our approach is a tailored viewpoint selection such that the content of each image can be fused into a seamless, textured 3D mesh. More specifically, we propose a continuous alignment strategy that iteratively fuses scene frames with the existing geometry.
|
|
StyleMesh: Style Transfer for Indoor 3D Scene Reconstructions
Lukas Höllein, Justin Johnson, Matthias Nießner
CVPR, 2022
project page
/
arXiv
/
video
/
code
We apply style transfer on mesh reconstructions of indoor scenes. We optimize an explicit texture for the reconstructed mesh of a scene and stylize it jointly from all available input images. Our depth- and angle-aware optimization leverages surface normal and depth data of the underlying mesh to create a uniform and consistent stylization for the whole scene.
|
Talks
|
TUMVision: Indoor Scene Generation From Diffusion Models, 24.10.2024, Munich
Google: Indoor Scene Generation From Diffusion Models, 14.10.2024, Munich
Voxel51: 3D-Consistent Image Generation with Text-to-Image Models, 06.06.2024, Online
|
Academic Service
|
Reviewer for CVPR2023, ICCV2023, SIGGRAPH ASIA 2023, CVPR2024, SIGGRAPH 2024, ECCV2024 (Outstanding Reviewer Award), CVPR2025, ICCV2025
|
Teaching
|
Advanced Deep Learning for Computer Vision: Visual Computing: Winter 24/25, Summer 25
advised 10+ students in semester-long course projects about neural rendering / novel-view-synthesis
3D Scanning & Motion Capture: Summer 22, Winter 22/23, Summer 23, Winter 23/24, Summer 24
advised 40+ students in 6-week course projects about stereo reconstruction, bundle adjustment, ARAP, and 3D reconstruction
Master's Thesis Supervision:
advised 4 students on topics about neural style transfer, world generation, and scene editing
|
Thank you Jon Barron for providing the source code of this website.
|
|