Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Feng, Zehui, Qiu, Tian, Wu, Tong, Li, Junxuan, Xu, Huayuan, Han, Ting
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2511.05393
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917067640274944
author Feng, Zehui
Qiu, Tian
Wu, Tong
Li, Junxuan
Xu, Huayuan
Han, Ting
author_facet Feng, Zehui
Qiu, Tian
Wu, Tong
Li, Junxuan
Xu, Huayuan
Han, Ting
contents Visual Quality Assessment (QA) seeks to predict human perceptual judgments of visual fidelity. While recent multimodal large language models (MLLMs) show promise in reasoning about image and video quality, existing approaches mainly rely on supervised fine-tuning or rank-only objectives, resulting in shallow reasoning, poor score calibration, and limited cross-domain generalization. We propose PreResQ-R1, a Preference-Response Disentangled Reinforcement Learning framework that unifies absolute score regression and relative ranking consistency within a single reasoning-driven optimization scheme. Unlike prior QA methods, PreResQ-R1 introduces a dual-branch reward formulation that separately models intra-sample response coherence and inter-sample preference alignment, optimized via Group Relative Policy Optimization (GRPO). This design encourages fine-grained, stable, and interpretable chain-of-thought reasoning about perceptual quality. To extend beyond static imagery, we further design a global-temporal and local-spatial data flow strategy for Video Quality Assessment. Remarkably, with reinforcement fine-tuning on only 6K images and 28K videos, PreResQ-R1 achieves state-of-the-art results across 10 IQA and 5 VQA benchmarks under both SRCC and PLCC metrics, surpassing by margins of 5.30% and textbf2.15% in IQA task, respectively. Beyond quantitative gains, it produces human-aligned reasoning traces that reveal the perceptual cues underlying quality judgments. Code and model are available.
format Preprint
id arxiv_https___arxiv_org_abs_2511_05393
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PreResQ-R1: Towards Fine-Grained Rank-and-Score Reinforcement Learning for Visual Quality Assessment via Preference-Response Disentangled Policy Optimization
Feng, Zehui
Qiu, Tian
Wu, Tong
Li, Junxuan
Xu, Huayuan
Han, Ting
Computer Vision and Pattern Recognition
Visual Quality Assessment (QA) seeks to predict human perceptual judgments of visual fidelity. While recent multimodal large language models (MLLMs) show promise in reasoning about image and video quality, existing approaches mainly rely on supervised fine-tuning or rank-only objectives, resulting in shallow reasoning, poor score calibration, and limited cross-domain generalization. We propose PreResQ-R1, a Preference-Response Disentangled Reinforcement Learning framework that unifies absolute score regression and relative ranking consistency within a single reasoning-driven optimization scheme. Unlike prior QA methods, PreResQ-R1 introduces a dual-branch reward formulation that separately models intra-sample response coherence and inter-sample preference alignment, optimized via Group Relative Policy Optimization (GRPO). This design encourages fine-grained, stable, and interpretable chain-of-thought reasoning about perceptual quality. To extend beyond static imagery, we further design a global-temporal and local-spatial data flow strategy for Video Quality Assessment. Remarkably, with reinforcement fine-tuning on only 6K images and 28K videos, PreResQ-R1 achieves state-of-the-art results across 10 IQA and 5 VQA benchmarks under both SRCC and PLCC metrics, surpassing by margins of 5.30% and textbf2.15% in IQA task, respectively. Beyond quantitative gains, it produces human-aligned reasoning traces that reveal the perceptual cues underlying quality judgments. Code and model are available.
title PreResQ-R1: Towards Fine-Grained Rank-and-Score Reinforcement Learning for Visual Quality Assessment via Preference-Response Disentangled Policy Optimization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.05393