Gated Fusion Network for Single Image Dehazing
Clone or download
Latest commit 92dfa9e Jul 11, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
inputs Add files via upload Mar 26, 2018
models Add files via upload Mar 26, 2018
results Add files via upload Mar 26, 2018
LoadSolver.m Add files via upload Mar 26, 2018
README.md Update README.md Jul 11, 2018
RealGWbal.m Add files via upload Mar 26, 2018
SolverParser.m Add files via upload Mar 26, 2018
caffe_init.m Add files via upload Mar 26, 2018
demo_test.m Add files via upload Mar 26, 2018
modelconfig_test.m Add files via upload Mar 26, 2018
psnr.m Add files via upload Mar 26, 2018
ssim.m Add files via upload Mar 26, 2018

README.md

Gated Fusion Network for Single Image Dehazing

Wenqi Ren, Lin Ma, Jiawei Zhang, Jinshan Pan, Xiaochun Cao, Wei Liu, Ming-Hsuan Yang

In this paper, we propose an efficient algorithm to directly restore a clear image from a hazy input. The proposed algorithm hinges on an end-to-end trainable neural network that consists of an encoder and a decoder. The encoder is exploited to capture the context of the derived input images, while the decoder is employed to estimate the contribution of each input to the final dehazed result using the learned representations attributed to the encoder. The constructed network adopts a novel fusion-based strategy which derives three inputs from an original hazy image by applying White Balance (WB), Contrast Enhancing (CE) and Gamma Correction (GC). We compute pixel-wise confidence maps based on the appearance differences between these different inputs to blend the information of the derived inputs and preserve the regions with pleasant visibility. The final dehazed image is yielded by gating the important features of the derived inputs. To train the network, we introduce a multi-scale based approach so that the halo artifacts can be avoided. Extensive experimental results on both synthetic and real-world images demonstrate that the proposed algorithm performs favorably against the stateof-the-art algorithms.

Setup

Please refer Caffe for installation details.
Basically, you need to first modify the MATLAB_DIR in Makefile.config and then run the demo_test.m for a successful compilation:

Scripts and pre-trained models

The pre-trained model can be found in models/snapshot.
Users can use demo_test to generate the results of any images.

Citations

Please cite this paper in your publications if it helps your research:
@inproceedings{Ren-CVPR-2018,
 author = {Ren, Wenqi and Ma, Lin and Zhang, Jiawei and Pan, Jinshan and Cao, Xiaochun and Liu, Wei and Yang, Ming-Hsuan},
 title = {Gated Fusion network for Single Image Dehazing},
 booktitle = {IEEE Conference on Computer Vision and Pattern Recognition},
 year = {2018}
}