• Stars
    star
    285
  • Rank 145,115 (Top 3 %)
  • Language
    C++
  • Created almost 7 years ago
  • Updated about 5 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Learning a Deep Single Image Contrast Enhancer from Multi-Exposure Images (TIP 2018)

Learning a Deep Single Image Contrast Enhancer from Multi-Exposure Images

Abstract

Due to the poor lighting condition and limited dynamic range of digital imaging devices, the recorded images are often under-/over-exposed and with low contrast. Most of previous single image contrast enhancement (SICE) methods adjust the tone curve to correct the contrast of an input image. Those methods, however, often fail in revealing image details because of the limited information in a single image. On the other hand, the SICE task can be better accomplished if we can learn extra information from appropriately collected training data. In this work, we propose to use the convolutional neural network (CNN) to train a SICE enhancer. One key issue is how to construct a training dataset of low-contrast and high-contrast image pairs for end-to-end CNN learning. To this end, we build a large-scale multi-exposure image dataset, which contains 589 elaborately selected high-resolution multi-exposure sequences with 4,413 images. Thirteen representative multi-exposure image fusion and stack-based high dynamic range imaging algorithms are employed to generate the contrast enhanced images for each sequence, and subjective experiments are conducted to screen the best quality one as the reference image of each scene. With the constructed dataset, a CNN can be easily trained as the SICE enhancer to improve the contrast of an under-/over-exposure image. Experimental results demonstrate the advantages of our method over existing SICE methods with a significant margin.

Code for training and testing

Trained Caffe model for the under-exposed image: *.caffemodel
Network structure: *.prototxt (to view the network structure, use this link)
Install and compile Caffe (the matlab interface is used)

Model 1 (End-to-end residual learning)

Run the Demo_Test.m for the result

image

Model 2 (Twostage Network)

Run the Demo_Test.m for the result

image

Model 3 (Twostage perpixel convolution)

Run the Demo_Test.m for the result

image

Dataset

Please refer to:

or

Requirements and Dependencies

Caffe

New Layers With CPU and GPU Implementations

caffe.proto (Parameters for SSIM and Regularization Layer)
Usage
layer {
  name: "SSIMLossLayer"
  type: "SSIMLoss"
  bottom: "output"
  bottom: "label"
  top: "SSIMLoss"
  ssim_loss_param{
    kernel_size: 8       
    stride: 8                
    c1: 0.0001              
    c2: 0.001                
  }
}

Citation

@article{Cai2018deep,
title={Learning a Deep Single Image Contrast Enhancer from Multi-Exposure Images}, 
author={Cai, Jianrui and Gu, Shuhang and Zhang, Lei},
journal={IEEE Transactions on Image Processing},
volume={27}, 
number={4}, 
pages={2049-2062}, 
year={2018}, 
publisher={IEEE}
}