DPC-VQA: Decoupling Quality Perception and Residual Calibration for Video Quality Assessment
This work addresses the high cost of adapting MLLMs to new VQA scenarios by proposing an efficient decoupling framework, offering a practical solution for resource-constrained settings.
DPC-VQA decouples quality perception from residual calibration for video quality assessment, using a frozen MLLM for perceptual priors and a lightweight branch for target-scenario adaptation. It achieves competitive performance with less than 2% trainable parameters and remains effective with only 20% of MOS labels.
Recent multimodal large language models (MLLMs) have shown promising performance on video quality assessment (VQA) tasks. However, adapting them to new scenarios remains expensive due to large-scale retraining and costly mean opinion score (MOS) annotations. In this paper, we argue that a pretrained MLLM already provides a useful perceptual prior for VQA, and that the main challenge is to efficiently calibrate this prior to the target MOS space. Based on this insight, we propose DPC-VQA, a decoupling perception and calibration framework for video quality assessment. Specifically, DPC-VQA uses a frozen MLLM to provide a base quality estimate and perceptual prior, and employs a lightweight calibration branch to predict a residual correction for target-scenario adaptation. This design avoids costly end-to-end retraining while maintaining reliable performance with lower training and data costs. Extensive experiments on both user-generated content (UGC) and AI-generated content (AIGC) benchmarks show that DPC-VQA achieves competitive performance against representative baselines, while using less than 2% of the trainable parameters of conventional MLLM-based VQA methods and remaining effective with only 20\% of MOS labels. The code will be released upon publication.