LaverNet: Lightweight All-in-one Video Restoration via Selective Propagation

TL;DR

LaverNet is a compact, lightweight video restoration network that effectively handles multiple degradations by selectively propagating degradation-agnostic features, achieving high performance with significantly fewer parameters.

Contribution

The paper introduces LaverNet, a small all-in-one video restoration model with a novel propagation mechanism for better temporal modeling under degradations.

Findings

LaverNet has only 362K parameters, less than 1% of existing models.

LaverNet achieves comparable or superior performance on benchmarks.

Selective feature propagation improves temporal modeling under degradations.

Abstract

Recent studies have explored all-in-one video restoration, which handles multiple degradations with a unified model. However, these approaches still face two challenges when dealing with time-varying degradations. First, the degradation can dominate temporal modeling, confusing the model to focus on artifacts rather than the video content. Second, current methods typically rely on large models to handle all-in-one restoration, concealing those underlying difficulties. To address these challenges, we propose a lightweight all-in-one video restoration network, LaverNet, with only 362K parameters. To mitigate the impact of degradations on temporal modeling, we introduce a novel propagation mechanism that selectively transmits only degradation-agnostic features across frames. Through LaverNet, we demonstrate that strong all-in-one restoration can be achieved with a compact network. Despite its small size, less than 1\% of the parameters of existing models, LaverNet achieves comparable, even superior performance across benchmarks.