Mitigating Content Shift and Hallucination in GenAI Image Editing via Structural Refinement
Quick Take
The proposed post-processing framework for GenAI image editing enhances structural fidelity while retaining perceptual quality, addressing issues like content hallucination. By establishing spatial correspondences, it effectively fuses input images with GenAI outputs, outperforming existing methods in aesthetic quality and pixel-level consistency.
Key Points
- Introduces 'structure-preserving GenAI fusion' for improved image retouching.
- Fuses input images with GenAI outputs to suppress hallucinated content.
- Demonstrates better aesthetic quality and pixel-level consistency in experiments.
- Evaluated against methods from photorealistic style transfer and image fusion.
- Addresses challenges faced by non-experts using GenAI image editors.
Article Content
From source RSS / original summaryarXiv:2605. 30437v1 Announce Type: new Abstract: Generative AI (GenAI) image editors, such as Nano Banana, produce visually compelling results for retouching tasks, enabling non-experts to edit images through text prompts alone. However, the generative nature of these models often introduces spatial misalignment, texture distortion, and content hallucination, all of which are detrimental to downstream workflows that require pixel-level fidelity.
We identify a problem setting we call "structure-preserving GenAI fusion" for black-box GenAI image retouching: retain the perceptual enhancements of a GenAI output while enforcing structural faithfulness to the original input image.
To address this problem, we propose a post-processing framework that fuses an input image with its GenAI-enhanced counterpart by first establishing coarse spatial and photometric correspondences, then performing a fusion stage that transfers desired enhancements while suppressing hallucinated content. In the absence of direct prior work in this setting, we evaluate our framework against representative methods from photorealistic style transfer and image fusion.
Our experiments demonstrate that our method better preserves aesthetic quality while maintaining pixel-level structural consistency and the input resolution.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
Evi-Steer introduces a novel evidential tuning framework for BiomedCLIP, enabling efficient fine-tuning with only 0.11% parameter updates. It significantly enhances performance in few-shot learning and domain shifts across 15 biomedical imaging datasets, demonstrating robustness for clinical applications.