merge_table_img.py

import cv2
import numpy as np
# ['baby', 'bird', 'butterfly', 'head', 'woman']
# ['baboon', 'comic', 'lenna', 'monarch', 'zebra']
for name in ['baby', 'bird', 'butterfly', 'head', 'woman']:
    img1 = cv2.imread('./results/psnr_pretrain/set5/Bic_SR_HR_%s.png' % name)
    img2 = cv2.imread('./results/esrgan/set5/Bic_SR_HR_%s.png' % name)
    padd = np.zeros([img1.shape[0], 5, 3], np.uint8)
    img_merge = np.concatenate((padd, img1, padd, img2, padd), 1)
    padd = np.zeros([5, img_merge.shape[1], 3], np.uint8)
    img_merge = np.concatenate((padd, img_merge, padd), 0)
    cv2.imwrite('./photo/table_%s.png' % name, img_merge)