ME-IQA: Memory-Enhanced Image Quality Assessment via Re-Ranking

TL;DR

ME-IQA enhances image quality assessment by integrating memory and re-ranking to produce more sensitive and continuous scores, overcoming the limitations of scalar score collapse in reasoning-based vision-language models.

Contribution

It introduces a novel memory-enhanced re-ranking framework that improves the sensitivity and continuity of IQA scores in reasoning-induced VLMs.

Findings

Consistent performance improvements across multiple IQA benchmarks.

Mitigation of discrete score collapse in reasoning-based models.

Effective use of memory and re-ranking for better quality assessment.

Abstract

Reasoning-induced vision-language models (VLMs) advance image quality assessment (IQA) with textual reasoning, yet their scalar scores often lack sensitivity and collapse to a few values, so-called discrete collapse. We introduce ME-IQA, a plug-and-play, test-time memory-enhanced re-ranking framework. It (i) builds a memory bank and retrieves semantically and perceptually aligned neighbors using reasoning summaries, (ii) reframes the VLM as a probabilistic comparator to obtain pairwise preference probabilities and fuse this ordinal evidence with the initial score under Thurstone's Case V model, and (iii) performs gated reflection and consolidates memory to improve future decisions. This yields denser, distortion-sensitive predictions and mitigates discrete collapse. Experiments across multiple IQA benchmarks show consistent gains over strong reasoning-induced VLM baselines, existing non-reasoning IQA methods, and test-time scaling alternatives.