CVMar 19, 2025

LEGION: Learning to Ground and Explain for Synthetic Image Detection

arXiv:2503.15264v156 citationsh-index: 16
Originality Incremental advance
AI Analysis

This addresses the problem of detecting AI-generated images for security and verification purposes, representing a domain-specific advancement with incremental improvements.

The paper tackles synthetic image detection by introducing SynthScars, a dataset with 12,236 synthetic images and fine-grained annotations, and LEGION, a multimodal framework that detects, segments, and explains artifacts, outperforming existing methods by 3.31% in mIoU and 7.75% in F1 score.

The rapid advancements in generative technology have emerged as a double-edged sword. While offering powerful tools that enhance convenience, they also pose significant social concerns. As defenders, current synthetic image detection methods often lack artifact-level textual interpretability and are overly focused on image manipulation detection, and current datasets usually suffer from outdated generators and a lack of fine-grained annotations. In this paper, we introduce SynthScars, a high-quality and diverse dataset consisting of 12,236 fully synthetic images with human-expert annotations. It features 4 distinct image content types, 3 categories of artifacts, and fine-grained annotations covering pixel-level segmentation, detailed textual explanations, and artifact category labels. Furthermore, we propose LEGION (LEarning to Ground and explain for Synthetic Image detectiON), a multimodal large language model (MLLM)-based image forgery analysis framework that integrates artifact detection, segmentation, and explanation. Building upon this capability, we further explore LEGION as a controller, integrating it into image refinement pipelines to guide the generation of higher-quality and more realistic images. Extensive experiments show that LEGION outperforms existing methods across multiple benchmarks, particularly surpassing the second-best traditional expert on SynthScars by 3.31% in mIoU and 7.75% in F1 score. Moreover, the refined images generated under its guidance exhibit stronger alignment with human preferences. The code, model, and dataset will be released.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes