11 papers
Unified Panoramic Geometry Estimation via Multi-View Foundation Models
Vukasin Bozic, Isidora Slavkovic, Dominik Narnhofer +4
Geometry estimation from perspective images has greatly advanced, maturing to the point where off-the-shelf foundation models are able to reconstruct 3D scene structure not only fr…
Understanding, Accelerating, and Improving MeanFlow Training
Jin-Young Kim, Hyojun Go, Lea Bogensperger +5
MeanFlow promises high-quality generative modeling in few steps, by jointly learning instantaneous and average velocity fields. Yet, the underlying training dynamics remain unclear…
Stitched Value Model for Diffusion Alignment
Hyojun Go, Hyungjin Chung, Prune Truong +8
For practical use, diffusion- or flow-based generative models must be aligned with task-specific rewards, such as prompt fidelity or aesthetic preference. That alignment is challen…
LoRA-Ensemble: Efficient Uncertainty Modelling for Self-Attention Networks
Dominik J. Mühlematter, Michelle Halbheer, Alexander Becker +4
Numerous real-world decisions rely on machine learning algorithms and require calibrated uncertainty estimates. However, modern methods often yield overconfident, uncalibrated pred…
Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator
Hyojun Go, Dominik Narnhofer, Goutam Bhat +3
The rapid progress of large, pretrained models for both visual content generation and 3D reconstruction opens up new possibilities for text-to-3D generation. Intuitively, one could…
Continuous Space-Time Video Super-Resolution with 3D Fourier Fields
Alexander Becker, Julius Erbach, Dominik Narnhofer +1
We introduce a novel formulation for continuous space-time video super-resolution. Instead of decoupling the representation of a video sequence into separate spatial and temporal c…