2015_Face_Detection Custom and custom48 layer

Hi Thank you for the code. I have two questions here, one is about custom and custom48 layers. Why do you define custom and custom48? they seem just convolutional layers to me. The second is about running on GPU. The running speed(3-4 seconds on large pic) is in GPU mode right? As long as the matconvnet is compiled in GPU mode, your code will run in GPU mode? Thank you for the time!

Mar 11 '16 01:03 Excaliburyz

@Excaliburyz
1.custom layer is to connect 12net fc layer to 24net. custom48 layer is to connect 24net fc layer to 48net. 2.I run code in Cpu mode, which cost 3-4 seconds. In gpu mode,it can be a bit faster but the bottleneck is get patch from picture. You can add "tic" "toc" to get accurate time cost.

Mar 11 '16 03:03 layumi

Thank you for the response! Still on the second question, if I compile the matconvnet in gpu mode, the code will run on GPU by default? So the gpu mode means only the cnn part is on GPU but other scripts are on CPU. Am I right?

Mar 12 '16 08:03 Excaliburyz

@Excaliburyz You can use "net = vl_simplenn_move(net,'gpu')" Yes.

Mar 13 '16 05:03 layumi