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

2026-04-19

AI Summary

This work addresses the prevalent issue of “edit spillover” in existing scene text editing methods, where non-target neighboring text is inadvertently altered. To mitigate this, the authors propose the Edit Fidelity Field (EFF), a training-free, plug-and-play mechanism that leverages OCR guidance to construct a four-region semantic structure and establishes a pixel-wise continuous fidelity field to explicitly protect non-target areas. Moving beyond conventional binary masks, EFF integrates diffusion model post-processing and introduces a region-wise spillover quantification protocol, substantially enhancing editing controllability. Experimental results demonstrate a dramatic reduction in edit spillover rate—from 94% to 25%—and a 91.4 dB improvement in PSNR for non-target region fidelity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Safety and Robustness
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

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.

Problem

• edit spillover
• scene text editing
• diffusion-based editing
• non-target region modification
• text preservation

Innovation

• Edit Fidelity Field
• scene text editing
• edit spillover
• diffusion-based editing
• training-free