Visual Distortion Detection in UGC Images Using Large Multimodal Models

The localized depiction of perceptual quality has long been a crucial, yet underexplored, challenge in image quality assessment (IQA). Existing approaches based on large multimodal models (LMMs) predominantly rely on text-driven supervised fine-tuning (SFT). However, this training paradigm exhibits notable limitations in detection accuracy. Moreover, synthetically distorted images, which are often used as the primary training data source, show a significant generalization gap when deployed in real-world scenarios; thus, the textbfsynthetic-to-authentic (textitS2A) problem represents a critical challenge. Motivated by these issues, we propose textbftextitVIGIL, which leverages the LMM architecture for precise visual distortion detection. From a candidate pool of over 1000K samples, we construct the textbftextitVIGIL-140K training set, which consists of over 140K distorted images. These images are obtained through rigorous quality filtering and carefully crafted distortion injection, covering 8 major synthetic distortion categories. Our model leverages different layers of the large language model (LLM) decoder, treating them as textitmultiple detectors that perform synchronous distortion detection using multi-level features. Additionally, we retain distortion cues from predictions assigned to the non-distortion class, which helps mitigate the ambiguous foreground-background (textitFG-BG) separation commonly encountered in the textitS2A problem. After post-processing, our model consistently outperforms strong baselines on both in-domain synthetic distortion detection and textitS2A tasks.