LIQA: Lifelong Blind Image Quality Assessment

TL;DR

The paper introduces LIQA, a lifelong learning approach for blind image quality assessment that adaptively learns new distortions without forgetting previous ones, suitable for real-world applications.

Contribution

LIQA is the first lifelong BIQA method that mitigates catastrophic forgetting and learns continuously from new distortions without access to past training data.

Findings

LIQA effectively handles continuous distortion shifts.

The model maintains stable performance over long task sequences.

It outperforms existing BIQA methods in lifelong learning scenarios.

Abstract

Existing blind image quality assessment (BIQA) methods are mostly designed in a disposable way and cannot evolve with unseen distortions adaptively, which greatly limits the deployment and application of BIQA models in real-world scenarios. To address this problem, we propose a novel Lifelong blind Image Quality Assessment (LIQA) approach, targeting to achieve the lifelong learning of BIQA. Without accessing to previous training data, our proposed LIQA can not only learn new distortions, but also mitigate the catastrophic forgetting of seen distortions. Specifically, we adopt the Split-and-Merge distillation strategy to train a single-head network that makes task-agnostic predictions. In the split stage, we first employ a distortion-specific generator to obtain the pseudo features of each seen distortion. Then, we use an auxiliary multi-head regression network to generate the predicted quality of each seen distortion. In the merge stage, we replay the pseudo features paired with pseudo labels to distill the knowledge of multiple heads, which can build the final regressed single head. Experimental results demonstrate that the proposed LIQA method can handle the continuous shifts of different distortion types and even datasets. More importantly, our LIQA model can achieve stable performance even if the task sequence is long.