Towards Explainable Partial-AIGC Image Quality Assessment

2025-04-12

AI Summary

This work addresses the underexplored problem of quality assessment for partially AI-generated (PAI) content. We formally define and tackle the challenge of interpretable quality evaluation for PAI images—a first in the literature. To this end, we introduce EPAIQA-15K, the first large-scale, explainable PAI quality assessment dataset, comprising 15K images and over 300K multidimensional human annotations. We propose a three-stage progressive training paradigm grounded in large vision-language models, unifying edit-region localization, quantitative scoring, and natural-language quality explanations. Our method jointly optimizes edit-region grounding, multidimensional perceptual modeling, and generative feedback. The released EPAIQA models significantly outperform baselines—achieving +18.7% improvement in Spearman rank correlation coefficient (SROCC) for quality scoring and +22.3% gain in BLEU-4 for explanation faithfulness. This work establishes a new paradigm for fine-grained, human-understandable governance of AIGC content.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Philosophy and Ethics of AI: Accountability, Interpretability & Explainability
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

The rapid advancement of AI-driven visual generation technologies has catalyzed significant breakthroughs in image manipulation, particularly in achieving photorealistic localized editing effects on natural scene images (NSIs). Despite extensive research on image quality assessment (IQA) for AI-generated images (AGIs), most studies focus on fully AI-generated outputs (e.g., text-to-image generation), leaving the quality assessment of partial-AIGC images (PAIs)-images with localized AI-driven edits an almost unprecedented field. Motivated by this gap, we construct the first large-scale PAI dataset towards explainable partial-AIGC image quality assessment (EPAIQA), the EPAIQA-15K, which includes 15K images with localized AI manipulation in different regions and over 300K multi-dimensional human ratings. Based on this, we leverage large multi-modal models (LMMs) and propose a three-stage model training paradigm. This paradigm progressively trains the LMM for editing region grounding, quantitative quality scoring, and quality explanation. Finally, we develop the EPAIQA series models, which possess explainable quality feedback capabilities. Our work represents a pioneering effort in the perceptual IQA field for comprehensive PAI quality assessment.

Problem

• Assessing quality of partially AI-edited natural images
• Lack of datasets for localized AI-driven edits evaluation
• Developing explainable models for multi-dimensional quality feedback

Innovation

• Constructed EPAIQA-15K dataset for partial-AIGC images
• Used large multi-modal models for three-stage training
• Developed explainable EPAIQA series models for quality feedback