The performance of existing image dehazing methods is limited by hand-designed features, such as the dark channel, color disparity and maximum contrast, with complex fusion schemes. In this paper, we propose a multi-scale deep neural network for single-image dehazing by learning the mapping between hazy images and their corresponding transmission maps. The proposed algorithm consists of a coarse-scale net which predicts a holistic transmission map based on the entire image, and a fine-scale net which refines results locally. To train the multiscale deep network, we synthesize a dataset comprised of hazy images and corresponding transmission maps based on the NYU Depth dataset. Extensive experiments demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods on both synthetic and real-world images in terms of quality and speed.
|Title of host publication||Computer Vision - 14th European Conference, ECCV 2016, Proceedings|
|Editors||Bastian Leibe, Nicu Sebe, Max Welling, Jiri Matas|
|Number of pages||16|
|Publication status||Published - 2016|
|Event||14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands|
Duration: 2016 Oct 8 → 2016 Oct 16
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||14th European Conference on Computer Vision, ECCV 2016|
|Period||16/10/8 → 16/10/16|
Bibliographical noteFunding Information:
This work is supported by National High-tech R&D Program of China (2014BAK11B03), National Basic Research Program of China (2013CB329305), National Natural Science Foundation of China (No. 61422213), “Strategic Priority Research Program” of the Chinese Academy of Sciences (XDA06010701), and National Program for Support of Top-notch Young Professionals. W. Ren is supported by a scholarship from China Scholarship Council. M.-H. Yang is supported in part by the NSF CAREER grant #1149783, and gifts from Adobe and Nvidia.
© Springer International Publishing AG 2016.
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)