Author

Fredrik Kahl

2 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Jun 2026

GeM-NR: Geometry-Aware Multi-View Editing for Nonrigid Scene Changes

GeM-NR is proposed, a fast and flexible training-free approach for general multi-view consistent image editing, including edits that drastically change the geometry and appearance of the scene, and is demonstrated to handle edits with significant changes in geometry and appearance.

Josef Bengtson, Yaroslava Lochman, Fredrik Kahl · arXiv.org · 1 citation

#artificial intelligence Preprint Nov 2025

3D-Consistent Multi-View Editing by Correspondence Guidance

Experimental results show that this approach significantly improves 3D consistency compared to existing multi-view editing methods, and enables high-quality Gaussian splat editing with sharp details and strong fidelity to user-specified text prompts.

Josef Bengtson, David Nilsson, Dong-In Lee et al. · 1 citation