Nighttime Hazy Image Enhancement via Progressively and Mutually Reinforcing Night-Haze Priors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Chen, Zhang, Huiwen, He, Mu, Li, Yujie, Qiao, Xiaotian
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914233769263104
author Zhu, Chen
Zhang, Huiwen
He, Mu
Li, Yujie
Qiao, Xiaotian
author_facet Zhu, Chen
Zhang, Huiwen
He, Mu
Li, Yujie
Qiao, Xiaotian
contents Enhancing the visibility of nighttime hazy images is challenging due to the complex degradation distributions. Existing methods mainly address a single type of degradation (e.g., haze or low-light) at a time, ignoring the interplay of different degradation types and resulting in limited visibility improvement. We observe that the domain knowledge shared between low-light and haze priors can be reinforced mutually for better visibility. Based on this key insight, in this paper, we propose a novel framework that enhances visibility in nighttime hazy images by reinforcing the intrinsic consistency between haze and low-light priors mutually and progressively. In particular, our model utilizes image-, patch-, and pixel-level experts that operate across visual and frequency domains to recover global scene structure, regional patterns, and fine-grained details progressively. A frequency-aware router is further introduced to adaptively guide the contribution of each expert, ensuring robust image restoration. Extensive experiments demonstrate the superior performance of our model on nighttime dehazing benchmarks both quantitatively and qualitatively. Moreover, we showcase the generalizability of our model in daytime dehazing and low-light enhancement tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2601_01998
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Nighttime Hazy Image Enhancement via Progressively and Mutually Reinforcing Night-Haze Priors
Zhu, Chen
Zhang, Huiwen
He, Mu
Li, Yujie
Qiao, Xiaotian
Computer Vision and Pattern Recognition
Enhancing the visibility of nighttime hazy images is challenging due to the complex degradation distributions. Existing methods mainly address a single type of degradation (e.g., haze or low-light) at a time, ignoring the interplay of different degradation types and resulting in limited visibility improvement. We observe that the domain knowledge shared between low-light and haze priors can be reinforced mutually for better visibility. Based on this key insight, in this paper, we propose a novel framework that enhances visibility in nighttime hazy images by reinforcing the intrinsic consistency between haze and low-light priors mutually and progressively. In particular, our model utilizes image-, patch-, and pixel-level experts that operate across visual and frequency domains to recover global scene structure, regional patterns, and fine-grained details progressively. A frequency-aware router is further introduced to adaptively guide the contribution of each expert, ensuring robust image restoration. Extensive experiments demonstrate the superior performance of our model on nighttime dehazing benchmarks both quantitatively and qualitatively. Moreover, we showcase the generalizability of our model in daytime dehazing and low-light enhancement tasks.
title Nighttime Hazy Image Enhancement via Progressively and Mutually Reinforcing Night-Haze Priors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.01998