About Global depth-wise convolution

[manmanCover]

@hujie-frank To my understanding, global convolution means to set the kernel size equals the current feature size, so this module can only be used on images which have the same sizes for training and testing? Is that right?

[GaoLiang Liu]

https://blog.csdn.net/hjimce/article/details/50187881
https://www.zhihu.com/question/270988169/answer/407731947

@manmanCover

[manmanCover]

https://blog.csdn.net/hjimce/article/details/50187881
https://www.zhihu.com/question/270988169/answer/407731947

@manmanCover

Sorry, 这两个链接没看懂，能帮忙解释一下吗？

[GaoLiang Liu]

训练和测试的输入尺寸确实是确定的。 可以通过vgg的dense的方法，来评估分类器。这样就能够用到全图的像素，而又固定输入尺寸。

…

---Original--- From: "manmanCover"<[e-mail]&gt; Date: Wed, Oct 30, 2019 00:01 AM To: "hujie-frank/GENet"<[e-mail]&gt;; Cc: "Comment"<[e-mail]&gt;;"GaoLiang Liu"<[e-mail]&gt;; Subject: Re: [hujie-frank/GENet] About Global depth-wise convolution (#3) https://blog.csdn.net/hjimce/article/details/50187881 https://www.zhihu.com/question/270988169/answer/407731947 @manmanCover Sorry, 这两个链接没看懂，能帮忙解释一下吗？ — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

[manmanCover]

训练和测试的输入尺寸确实是确定的。 可以通过vgg的dense的方法，来评估分类器。这样就能够用到全图的像素，而又固定输入尺寸。
…
---Original--- From: "manmanCover"<[e-mail]> Date: Wed, Oct 30, 2019 00:01 AM To: "hujie-frank/GENet"<[e-mail]>; Cc: "Comment"<[e-mail]>;"GaoLiang Liu"<[e-mail]>; Subject: Re: [hujie-frank/GENet] About Global depth-wise convolution (#3) https://blog.csdn.net/hjimce/article/details/50187881 https://www.zhihu.com/question/270988169/answer/407731947 @manmanCover Sorry, 这两个链接没看懂，能帮忙解释一下吗？ — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

懂啦！谢！