Gated Fusion Network for Single Image Dehazing
Wenqi Ren1,2, Lin Ma2, Jiawei Zhang3, Jinshan Pan4, Xiaochun Cao1, Wei Liu2 and Ming-Hsuan Yang3
1IIE CAS, 2Tecent AI Lab, 3UCMerced, 4Nanjing University of Science and Technology


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 state-of-the-art algorithms.


Figure 1. Image dehazing result. We exploit a gated fusion network for single image deblurring. (a) Hazy input. (b)-(d) are the derived inputs. (f)-(h) are learned confidence maps for (b), (c) and (d) respectively. (e) Our result.


More results




Please cite this paper in your publications if it helps your research:
   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}