ResViT 代码复现和讲解

论文题目：ResViT: Residual vision transformers for multi-modal medical image synthesis

论文地址：[2106.16031] ResViT: Residual vision transformers for multi-modal medical image synthesis (arxiv.org)

项目地址：GitHub - icon-lab/ResViT: Official Implementation of ResViT: Residual Vision Transformers for Multi-modal Medical Image Synthesis

一种新的用于多模态医学图像合成的生成对抗方法

复现

1、从github下载源码 

2、数据问题：

本次实验使用的数据不同于原项目，因此给出了新的dataset 方法和预处理方法

BraTS2020 Dataset (Training + Validation)

BraTS2020 Dataset (Training + Validation) (kaggle.com)

下载后解压

3、预处理，

预处理方法使用的MTNet的项目方法，代码讲解可以看我之前的博客

import numpy as np
from matplotlib import pylab as plt
import nibabel as nib
import random
import glob
import os
from PIL import Image
import imageio

def normalize(image, mask=None, percentile_lower=0.2, percentile_upper=99.8):

    if mask is None:
        mask = image