iDETEX: Empowering MLLMs for Intelligent DETailed EXplainable IQA
This work addresses the need for interpretable image quality assessment for applications in computer vision and multimedia, representing an incremental advancement by integrating existing methods into a unified model.
The paper tackles the challenge of detailed and explainable image quality assessment by proposing iDETEX, a multimodal large language model that simultaneously performs quality grounding, perception, and description, achieving state-of-the-art performance on the ViDA-UGC benchmark and ranking first in the ICCV MIPI 2025 challenge.
Image Quality Assessment (IQA) has progressed from scalar quality prediction to more interpretable, human-aligned evaluation paradigms. In this work, we address the emerging challenge of detailed and explainable IQA by proposing iDETEX-a unified multimodal large language model (MLLM) capable of simultaneously performing three key tasks: quality grounding, perception, and description. To facilitate efficient and generalizable training across these heterogeneous subtasks, we design a suite of task-specific offline augmentation modules and a data mixing strategy. These are further complemented by online enhancement strategies to fully exploit multi-sourced supervision. We validate our approach on the large-scale ViDA-UGC benchmark, where iDETEX achieves state-of-the-art performance across all subtasks. Our model ranks first in the ICCV MIPI 2025 Detailed Image Quality Assessment Challenge, demonstrating its effectiveness and robustness in delivering accurate and interpretable quality assessments.