VLM-Augmented Degradation Modeling for Image Restoration Under Adverse Weather Conditions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shao, Qianyi, Zhang, Yuanfan, Xiao, Renxiang, Hu, Liang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909915998584832
author Shao, Qianyi
Zhang, Yuanfan
Xiao, Renxiang
Hu, Liang
author_facet Shao, Qianyi
Zhang, Yuanfan
Xiao, Renxiang
Hu, Liang
contents Reliable visual perception under adverse weather conditions, such as rain, haze, snow, or a mixture of them, is desirable yet challenging for autonomous driving and outdoor robots. In this paper, we propose a unified Memory-Enhanced Visual-Language Recovery (MVLR) model that restores images from different degradation levels under various weather conditions. MVLR couples a lightweight encoder-decoder backbone with a Visual-Language Model (VLM) and an Implicit Memory Bank (IMB). The VLM performs chain-of-thought inference to encode weather degradation priors and the IMB stores continuous latent representations of degradation patterns. The VLM-generated priors query the IMB to retrieve fine-grained degradation prototypes. These prototypes are then adaptively fused with multi-scale visual features via dynamic cross-attention mechanisms, enhancing restoration accuracy while maintaining computational efficiency. Extensive experiments on four severe-weather benchmarks show that MVLR surpasses single-branch and Mixture-of-Experts baselines in terms of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). These results indicate that MVLR offers a practical balance between model compactness and expressiveness for real-time deployment in diverse outdoor conditions.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16998
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VLM-Augmented Degradation Modeling for Image Restoration Under Adverse Weather Conditions
Shao, Qianyi
Zhang, Yuanfan
Xiao, Renxiang
Hu, Liang
Computer Vision and Pattern Recognition
Reliable visual perception under adverse weather conditions, such as rain, haze, snow, or a mixture of them, is desirable yet challenging for autonomous driving and outdoor robots. In this paper, we propose a unified Memory-Enhanced Visual-Language Recovery (MVLR) model that restores images from different degradation levels under various weather conditions. MVLR couples a lightweight encoder-decoder backbone with a Visual-Language Model (VLM) and an Implicit Memory Bank (IMB). The VLM performs chain-of-thought inference to encode weather degradation priors and the IMB stores continuous latent representations of degradation patterns. The VLM-generated priors query the IMB to retrieve fine-grained degradation prototypes. These prototypes are then adaptively fused with multi-scale visual features via dynamic cross-attention mechanisms, enhancing restoration accuracy while maintaining computational efficiency. Extensive experiments on four severe-weather benchmarks show that MVLR surpasses single-branch and Mixture-of-Experts baselines in terms of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM). These results indicate that MVLR offers a practical balance between model compactness and expressiveness for real-time deployment in diverse outdoor conditions.
title VLM-Augmented Degradation Modeling for Image Restoration Under Adverse Weather Conditions
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.16998