Q-Router：基于专家模型路由与伪影定位的智能体视频质量评估
Q-Router: Agentic Video Quality Assessment with Expert Model Routing and Artifact Localization
AI总结：
针对现有视频质量评估模型泛化差、可解释性弱等问题，提出智能体框架Q-Router，通过VLM实时路由集成专家模型并定位时空伪影，显著提升了性能、泛化性与可解释性。
摘要：
视频质量评估（VQA）是一项基础的计算机视觉任务，旨在预测给定视频的感知质量，使其与人类判断保持一致。使用直接分数监督训练的现有高性能VQA模型存在以下问题：（1）在跨用户生成内容（UGC）、短视频以及AI生成内容（AIGC）等多样内容和任务上泛化能力差；（2）可解释性有限；（3）缺乏对新用例或内容类型的可扩展性。我们提出了Q-Router，一个具有多层模型路由系统的通用VQA智能体框架。Q-Router集成了多种专家模型，并采用视觉语言模型（VLM）作为实时路由器，根据输入视频语义动态推理并集成最合适的专家。我们基于计算预算构建了多层路由系统，其中最重层级包含用于提升可解释性的特定时空伪影定位。这种智能体设计使Q-Router能结合专业专家的互补优势，在异构视频源和任务上保持一致性能，实现灵活性与鲁棒性。大量实验表明，Q-Router在多个基准上匹配或超越最先进的VQA模型，同时显著提升了泛化能力和可解释性。此外，Q-Router在基于质量的问答基准Q-Bench-Video上表现优异，突显其作为下一代VQA系统基础的潜力。最后，我们证明Q-Router能有效定位时空伪影，展现出作为后训练视频生成模型奖励函数的潜力。
英文摘要：
Video quality assessment (VQA) is a fundamental computer vision task that aims to predict the perceptual quality of a given video in alignment with human judgments. Existing performant VQA models trained with direct score supervision suffer from (1) poor generalization across diverse content and tasks, ranging from user-generated content (UGC), short-form videos, to AI-generated content (AIGC), (2) limited interpretability, and (3) lack of extensibility to novel use cases or content types. We propose Q-Router, an agentic framework for universal VQA with a multi-tier model routing system. Q-Router integrates a diverse set of expert models and employs vision--language models (VLMs) as real-time routers that dynamically reason and then ensemble the most appropriate experts conditioned on the input video semantics. We build a multi-tiered routing system based on the computing budget, with the heaviest tier involving a specific spatiotemporal artifacts localization for interpretability. This agentic design enables Q-Router to combine the complementary strengths of specialized experts, achieving both flexibility and robustness in delivering consistent performance across heterogeneous video sources and tasks. Extensive experiments demonstrate that Q-Router matches or surpasses state-of-the-art VQA models on a variety of benchmarks, while substantially improving generalization and interpretability. Moreover, Q-Router excels on the quality-based question answering benchmark, Q-Bench-Video, highlighting its promise as a foundation for next-generation VQA systems. Finally, we show that Q-Router capably localizes spatiotemporal artifacts, showing potential as a reward function for post-training video generation models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
