WM-MoE: Weather-aware Multi-scale Mixture-of-Experts for Blind Adverse Weather Removal
Yulin Luo, Rui Zhao, Xiaobao Wei, Jinwei Chen, Yijie Lu, Shenghao Xie,, Tianyu Wang, Ruiqin Xiong, Ming Lu, Shanghang Zhang

TL;DR
This paper introduces WM-MoE, a Transformer-based model with weather-aware routing and multi-scale features, achieving state-of-the-art results in blind adverse weather removal for autonomous driving scenarios.
Contribution
The paper proposes WM-MoE with WEAR and MSE components, and Weather Guidance Fine-grained Contrastive Learning, to effectively handle unknown weather types and intensities.
Findings
Achieves state-of-the-art performance on public datasets.
Effectively handles multiple adverse weather types.
Improves downstream segmentation tasks.
Abstract
Adverse weather removal tasks like deraining, desnowing, and dehazing are usually treated as separate tasks. However, in practical autonomous driving scenarios, the type, intensity,and mixing degree of weather are unknown, so handling each task separately cannot deal with the complex practical scenarios. In this paper, we study the blind adverse weather removal problem. Mixture-of-Experts (MoE) is a popular model that adopts a learnable gate to route the input to different expert networks. The principle of MoE involves using adaptive networks to process different types of unknown inputs. Therefore, MoE has great potential for blind adverse weather removal. However, the original MoE module is inadequate for coupled multiple weather types and fails to utilize multi-scale features for better performance. To this end, we propose a method called Weather-aware Multi-scale MoE (WM-MoE) based…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFlood Risk Assessment and Management · Precipitation Measurement and Analysis
MethodsMulti-Head Attention · Attention Is All You Need · Linear Layer · Adam · Absolute Position Encodings · Softmax · Residual Connection · Byte Pair Encoding · Label Smoothing · Position-Wise Feed-Forward Layer
