期刊
IEEE MULTIMEDIA
卷 28, 期 3, 页码 97-106出版社
IEEE COMPUTER SOC
DOI: 10.1109/MMUL.2021.3052821
关键词
Image restoration; Lighting; Feature extraction; Convolution; Atmospheric modeling; Two dimensional displays; Spectral analysis; Single image dehazing; two-shot network; region-adaptive fusion
类别
资金
- National Natural Science Foundation of China [61971095, 61871078, 61831005, 61871087]
The study proposes a region-adaptive two-shot network (RATNet) dehazing algorithm that optimizes filtering effects for different regions through a coarse-to-fine framework. Experimental results demonstrate the superior dehazing performance of the algorithm.
Single image dehazing is the key to enhancing image visibility in outdoor scenes, which facilitates human observation and computer recognition. The existing approaches generally utilize a one-shot strategy that indiscriminately applies the same filters to all local regions. However, due to neglecting inhomogeneous illumination and detail distortion, their dehazed results easily suffer from underfiltering or overfiltering across different regions. To tackle this issue, we propose a region-adaptive two-shot network (RATNet) that follows a coarse-to-fine framework. First, a lightweight subnetwork is applied to execute regular global filtering and obtain an initially restored image. Then, a two-branch subnetwork is put forward whose branches separately refine its illumination and detail. Eventually, we derive the final prediction by adaptively aggregating the results after illumination modification and detail restoration, whose region-variant weights are jointly optimized by maximizing the similarity between our fused result and haze-free counterpart. Extensive experiments validate the superiority of our proposed algorithm.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据