Chongyi Li

8块 32G V100 FireWallDragonDarkFluid ***@***.***> 于2023年7月1日周六 12:01写道： > How many GPUs are used during training? > In the paper, it is mentioned a single V100 is used for implement, >...

Such an image with obvious noise challenges all low-light image enhancement methods. Zero-DCE can only enhance low-light image without obvious artifact or noise. Unfortunately, it cannot handle artifacts, blocking, and...

MEITU is commercial software that contains a serial of SP operations, not just light enhancement. Moreover, I guess that it operates on raw data (not sure). So it is unfair...

Hi, Thanks for asking. The company has several methods to enhance lowlight image and denoising with Zero-DCE coupled with a denoising technique. However, I cannot provide the details because of...

是的。 a5160533 ***@***.***> 于2022年8月5日周五 17:24写道： > 您好! 請問為何是local region 大小是4x4, 而不是3x3或是5x5呢? > 是因為輸入圖片都resize 成256x256, 怕沒辦法算到每一塊嗎? > 感謝! > > — > Reply to this email directly, view it on GitHub...

方法对大小那么敏感 所 8*8 也可以。 a5160533 ***@***.***> 于2022年8月5日周五 19:37写道： > 抱歉!我想再請問一下,那為何不能調成8x8, 16x16呢? > > — > Reply to this email directly, view it on GitHub > , > or unsubscribe >...

是的 a5160533 ***@***.***> 于2022年8月5日周五 20:48写道： > 是不敏感所以都可以的意思嗎?感謝! > > — > Reply to this email directly, view it on GitHub > , > or unsubscribe > > . > You...

> 您好,我想請問為何在model.py的程式碼裡面,在原本每某些層的後面有註解掉的 maxpooling 跟 upsampling呢? 这是因为，我们后面发现下采样和上采样会破坏曲线参数图相邻像素的约束关系。

由于是假期，南开的服务器外部借口必须关闭，等开学会自动开放。感谢关注~ Smile-QT ***@***.***> 于2023年8月3日周四 10:53写道： > 你好，请问你解决了VPN问题吗？ > > — > Reply to this email directly, view it on GitHub > , > or unsubscribe > > . > You...

Hi, You can manually set the patch size in the different training stages, ranging from small to large. We found this kind of training strick can improve the performance. It...