Research map / Image generation / Editing and style transfer
Training-Free 3D Editing via Feature-Divergence Localization and Trajectory Correction
Tri Ton, Siwoo Lim, Joshua Tian Jin Tee, Chang Yoo
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a training-free 3D editing method using feature-divergence localization and trajectory correction on probability flow models, aiming to edit targeted regions while preserving unedited geometry.
Problem
Targeted structural editing of 3D assets while strictly preserving the geometry of unedited regions; existing inversion-free flow trajectories use global velocity updates that cause spatially unintended changes.
Approach
Localizes edit regions via feature divergence between source and target trajectories and applies trajectory correction to confine modifications within a latent space, avoiding global velocity updates.
Closest papers by meaning
- Velocity-Space 3D Asset Editing (NeurIPS 2026 · similarity 0.89)
- AnchorFlow: Training-Free 3D Editing via Latent Anchor-Aligned Flows (CVPR 2026 · similarity 0.88)
- PrEditor3D: Fast and Precise 3D Shape Editing (CVPR 2025 · similarity 0.86)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.86)
- CTRL-D: Controllable Dynamic 3D Scene Editing with Personalized 2D Diffusion (CVPR 2025 · similarity 0.86)
- FocalDreamer: Text-Driven 3D Editing via Focal-Fusion Assembly (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
