UniRain: Unified Image Deraining with RAG-based Dataset Distillation and Multi-objective Reweighted Optimization

TL;DR

UniRain introduces a universal image deraining framework that generalizes across various rain types and conditions by combining dataset distillation, RAG-based sample selection, and a multi-objective optimization within an MoE architecture.

Contribution

The paper presents a novel unified deraining model with a RAG-based dataset distillation pipeline and a multi-objective reweighted optimization strategy for improved generalization.

Findings

Outperforms state-of-the-art models on multiple datasets

Effectively handles diverse rain degradations and conditions

Enhances robustness and generalization in image deraining

Abstract

Despite significant progress has been made in image deraining, we note that most existing methods are often developed for only specific types of rain degradation and fail to generalize across diverse real-world rainy scenes. How to effectively model different rain degradations within a universal framework is important for real-world image deraining. In this paper, we propose UniRain, an effective unified image deraining framework capable of restoring images degraded by rain streak and raindrop under both daytime and nighttime conditions. To better enhance unified model generalization, we construct an intelligent retrieval augmented generation (RAG)-based dataset distillation pipeline that selects high-quality training samples from all public deraining datasets for better mixed training. Furthermore, we incorporate a simple yet effective multi-objective reweighted optimization strategy into the asymmetric mixture-of-experts (MoE) architecture to facilitate consistent performance and improve robustness across diverse scenes. Extensive experiments show that our framework performs favorably against the state-of-the-art models on our proposed benchmarks and multiple public datasets.