README.md 1.1 KB
Newer Older
yongshk's avatar
更新  
yongshk committed
1
2
3
4
5
6
7
# LapSRN
## 模型介绍
LapSRN是一种用于超分辨率重建的神经网络模型。它可以将低分辨率的图像增强到高分辨率,从而提高图像的质量和清晰度。
## 模型结构
LapSRN模型主要有两个部分,即拉普拉斯金字塔预测模型和残差学习模型。
## 数据集
数据集下载地址:DIV2K
yongshk's avatar
yongshk committed
8

yongshk's avatar
更新  
yongshk committed
9
https://data.vision.ee.ethz.ch/cvl/DIV2K/
yongshk's avatar
yongshk committed
10

yongshk's avatar
更新  
yongshk committed
11
12
## 训练及推理
### 环境配置
yongshk's avatar
yongshk committed
13

yongshk's avatar
更新  
yongshk committed
14
加载环境变量
yongshk's avatar
yongshk committed
15
16
17
18
19

export PATH={PYTHON3_install_dir}/bin:$PATH

export LD_LIBRARY_PATH={PYTHON3_install_dir}/lib:$LD_LIBRARY_PATH

yongshk's avatar
更新  
yongshk committed
20
python依赖安装:
yongshk's avatar
yongshk committed
21

yongshk's avatar
更新  
yongshk committed
22
23
24
    tensorflow==1.15.1
### 训练
训练命令:
yongshk's avatar
yongshk committed
25

yongshk's avatar
更新  
yongshk committed
26
27
28
29
    config.valid.hr_folder_path = '/../LapSRN/data/DIV2K_valid_HR/'
    config.valid.lr_folder_path = '/../LapSRN/data/DIV2K_train_LR_bicubic/X4/'
    
    python main.py
yongshk's avatar
yongshk committed
30
31


yongshk's avatar
更新  
yongshk committed
32
33
### 测试
测试命令:
yongshk's avatar
yongshk committed
34

yongshk's avatar
更新  
yongshk committed
35
36
    python main.py -m test \
                   -f TESTIMAGE
yongshk's avatar
yongshk committed
37
38
39
40




yongshk's avatar
更新  
yongshk committed
41
42
43
44
## 源码仓库及问题反馈
* [https://github.com/zjuela/LapSRN-tensorflow](https://github.com/zjuela/LapSRN-tensorflow)
## 参考
* [https://github.com/zjuela/LapSRN-tensorflow](https://github.com/zjuela/LapSRN-tensorflow)