AI Paper Atlas

Research map / Image generation / Editing and style transfer

TopoRefine: Topology-Aware Correspondence and Residual Refinement for Training-Free Subject-Consistent Generation

Zhanxin Gao, Zexin Ti, Chen Zhao, Beier Zhu, Xiantao Hu, Yanhao Ge, Jian Yang, Ying Tai

NeurIPS 2026 · Image generation › Editing and style transfer · approach: Texture and sketch synthesis

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 TopoRefine, a training-free method for subject-consistent generation that uses topology-aware correspondence and residual refinement to transfer subject identity without structural mismatch.

Problem

Subject-consistent generation must preserve subject identity across diverse text-driven contexts; semantic-driven correspondence can transfer identity features to appearance-similar but structurally incompatible regions, and extended attention can cause issues.

Approach

Replaces purely semantic correspondence with topology-aware correspondence for identity transfer and adds residual refinement, in a training-free manner (details beyond the abstract are not stated in the provided text).

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy