/Multi-scale-CNN-Dehazing

Single Image Dehazing via Multi-scale Convolutional Neural Networks, ECCV 2016

Primary LanguageCuda

ECCV-Single image dehazing via Multi-scale Convolutional Neural Networks

Wenqi Ren, Si Liu, Hua Zhang, Jinshan Pan, Xiaochun Cao, and Ming-Hsuan Yang

Released in 10/10/2016. The codes are based on MatConvNet.

Description

This is a test implementation for ECCV16 paper: Single Image Dehazing via Multi-scale Convolutional Neural Networks. We provide testing demo as well as the pretrained model.

Scripts and pre-trained model

This is a demo implementation for our dehazing paper in ECCV16.

This test code is based on the MatConvNet toolbox. You should compile MatConvNet first on your computer then use the "demo_MSCNNdehazing.m" to test on your images.

The current published compliled version in the folder "MSCNN_dehazing\matlab" can be directly use on a Win7 computer.

Citations

  • @inproceedings{Ren-ECCV-2016,
  •    author = {Ren, Wenqi and Liu, Si and Zhang, Hua and Pan, Jinshan and Cao, Xiaochun and Yang, Ming-Hsuan},
  •    title = {Single Image Dehazing via Multi-Scale Convolutional Neural Networks},
  •    booktitle = {European Conference on Computer Vision},
  •    year = {2016}
  • }