Edit Fidelity Field: Semantics-Aware Region Isolation for Training-Free Scene Text Editing

TL;DR

This paper introduces the Edit Fidelity Field (EFF), a semantics-aware post-processing module that significantly reduces spillover in diffusion-based scene text editing, improving non-target region preservation.

Contribution

The paper proposes EFF, a training-free, model-agnostic method that controls pixel-level editing fidelity using OCR-detected regions, addressing spillover in scene text editing.

Findings

EFF reduces spillover rate from 94% to 25%.

Non-target region preservation improves by +91.4 dB PSNR.

The method is applicable to any diffusion-based scene text editing model.

Abstract

Scene text editing (STE) has achieved remarkable progress in accurately rendering target text through diffusion-based methods. However, we identify a critical yet overlooked problem: edit spillover -- when editing a target text region, existing methods inadvertently modify non-target regions, particularly neighboring text. Through systematic evaluation on 50 real-world scenes across four categories, we reveal that state-of-the-art diffusion editing models exhibit a spillover rate of 94%, meaning nearly all non-target text regions are altered during editing. To address this, we propose the Edit Fidelity Field (EFF), a semantics-aware continuous field that controls per-pixel editing fidelity. Unlike binary masks, EFF leverages OCR-detected text regions to construct a four-zone field: Edit Core (fully editable), Transition Zone (smooth decay), Protected Zone (non-target text, explicitly locked), and Background (strictly preserved). EFF operates as a training-free, model-agnostic post-processing module applicable to any diffusion-based STE method. We further propose per-region spillover quantification, a novel evaluation protocol that measures edit leakage at each non-target text region individually. Experiments demonstrate that EFF reduces spillover rate from 94% to 25% while improving non-target region preservation by +91.4 dB PSNR.