同步操作将从 WinterDream/Super_resolution 强制同步,此操作会覆盖自 Fork 仓库以来所做的任何修改,且无法恢复!!!
确定后同步将在后台操作,完成时将刷新页面,请耐心等待。
This repository is implementation of the "Image Super-Resolution Using Very Deep Residual Channel Attention Networks".
Tensorflow is required for quickly fetching image in training phase.
For below results, we set the number of residual groups as 6, the number of RCAB as 12, the number of features as 64.
In addition, we use a intermediate weights because training process need to take a looong time on my computer. 😭
Original | BICUBIC x2 | RCAN x2 |
When training begins, the model weights will be saved every epoch.
If you want to train quickly, you should use --use_fast_loader option.
Output results consist of restored images by the BICUBIC and the RCAN.
比赛最后生成的图片像素有点大,资源有限,自己掏钱买了一台1080ti电脑,吃了大半年土,奈何一块1080ti还是太low。 对一块显卡的人来说DBPN,RCAN层数必须要减少一下,否则测试会内存溢出,也试了下cvpr2019的SAN,也必须阉割一下。后面由于实习等原因放弃了,能学以致,这次比赛真的是不错的体验,一起加油吧
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。