Zoom-IQA: Image Quality Assessment with Reliable Region-Aware Reasoning

TL;DR

Zoom-IQA is a novel vision language model for image quality assessment that emphasizes reliable region-aware reasoning, uncertainty awareness, and iterative refinement to improve robustness, explainability, and generalization.

Contribution

It introduces a two-stage training pipeline with supervised fine-tuning and reinforcement learning, enhancing reasoning reliability and scoring accuracy in IQA.

Findings

Achieves improved robustness and explainability in IQA

Demonstrates better generalization across datasets

Effective in downstream tasks like image restoration

Abstract

Image Quality Assessment (IQA) is a long-standing problem in computer vision. Previous methods typically focus on predicting numerical scores without explanation or providing low-level descriptions lacking precise scores. Recent reasoning-based vision language models (VLMs) have shown strong potential for IQA by jointly generating quality descriptions and scores. However, existing VLM-based IQA methods often suffer from unreliable reasoning due to their limited capability of integrating visual and textual cues. In this work, we introduce Zoom-IQA, a VLM-based IQA model to explicitly emulate key cognitive behaviors: uncertainty awareness, region reasoning, and iterative refinement. Specifically, we present a two-stage training pipeline: 1) supervised fine-tuning (SFT) on our Grounded-Rationale-IQA (GR-IQA) dataset to teach the model to ground its assessments in key regions, and 2) reinforcement learning (RL) for dynamic policy exploration, stabilized by our KL-Coverage regularizer to prevent reasoning and scoring diversity collapse, with a Progressive Re-sampling Strategy for mitigating annotation bias. Extensive experiments show that Zoom-IQA achieves improved robustness, explainability, and generalization. The application to downstream tasks, such as image restoration, further demonstrates the effectiveness of Zoom-IQA.