Get Started
Home
Topics
Search
Library
Research questionHow can multi-view image editing preserve geometric and photometric consistency across views of a 3D scene?Editing each view independently can make corresponding scene points change differently in position or appearance. These disagreements undermine the consistency of reconstructed or editable 3D scene representations.
AI
Computer Vision
Diffusion Models
Image & Video Processing
Image Generation
Machine Learning
Latest papersRecent research connected to this question, newest first.3D-Consistent Multi-View Editing by Correspondence GuidanceThe source concerns text-based image editing with diffusion and flow models, with implications for NeRF and Gaussian splat editing. It reports a training-free guidance framework evaluated with dense and sparse multi-view setups and combined with varied image-editing methods.research paper · Sep 1, 2026
Related questions
How can instruction-guided 3D editing avoid geometric artifacts and cross-view inconsistency without paired training data?How can multi-view 3D reconstruction predict precise depth while preserving controllable uncertainty over plausible shapes?How can point-based 3D representations support coordinated, fine-grained geometry and texture editing?How can sparse-view novel-view synthesis maintain geometric and temporal consistency along a camera trajectory?