Multi channel residual network model for accurate estimation of spatially-varying and depth-dependent defocus kernels
By Yanpeng Cao, Zhangyu Ye, Zewei He, Jiangxin Yang, Yanlong Cao, Christel-Loic Tisse, and Michael Ying Yang.
The paper is available at |[PDF Download]
In this paper, we propose a novel multi-channel residual deep network model to learn the end-to-end mapping function between the in-focus and out-of-focus image patches captured at different spatial locations and depths.Besides, we construct a dataset containing well-aligned in-focus, out-of-focus and depth image.
The source image data come from DIV2K: https://data.vision.ee.ethz.ch/cvl/DIV2K/
The dataset contributed in the paper can be download at:
- Baidu Cloud: https://pan.baidu.com/s/1f6IcpXHaXiskxoCDBIzY0w Password：xemh
This code is based on Caffe. Thanks to the contributors of Caffe.
Caffe: https://github.com/BVLC/caffe
If you have any questions, feel free to contact:
- Yanpeng Cao (caoyp@zju.edu.cn)
- Zhangyu Ye (yezhangyu@yeah.net)
Copyright(c) Yanpeng Cao and Zhangyu Ye All rights reserved.