Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing

TL;DR

Q-Probe introduces a novel context-aware agentic framework for high-resolution image quality assessment, overcoming limitations of existing models by capturing subtle local degradations and eliminating biases.

Contribution

It presents the first agentic IQA framework with a new benchmark and a three-stage training paradigm for fine-grained, high-resolution image quality evaluation.

Findings

Q-Probe achieves state-of-the-art performance in high-resolution IQA.

The framework effectively captures subtle local degradations.

It maintains efficacy across different resolution scales.

Abstract

Reinforcement Learning (RL) has empowered Multimodal Large Language Models (MLLMs) to achieve superior human preference alignment in Image Quality Assessment (IQA). However, existing RL-based IQA models typically rely on coarse-grained global views, failing to capture subtle local degradations in high-resolution scenarios. While emerging "Thinking with Images" paradigms enable multi-scale visual perception via zoom-in mechanisms, their direct adaptation to IQA induces spurious "cropping-implies-degradation" biases and misinterprets natural depth-of-field as artifacts. To address these challenges, we propose Q-Probe, the first agentic IQA framework designed to scale IQA to high resolution via context-aware probing. First, we construct Vista-Bench, a pioneering benchmark tailored for fine-grained local degradation analysis in high-resolution IQA settings. Furthermore, we propose a three-stage training paradigm that progressively aligns the model with human preferences, while simultaneously eliminating causal bias through a novel context-aware cropping strategy. Extensive experiments demonstrate that Q-Probe achieves state-of-the-art performance in high-resolution settings while maintaining superior efficacy across resolution scales.