search

EdwardTyantov / ultrasound-nerve-segmentation

Kaggle Ultrasound Nerve Segmentation competition [Keras]
181 stars 61 forks source link

upsampling2D and deconvolution2D #10

Closed wenouyang closed 7 years ago

wenouyang commented 7 years ago

Hi Edward,

In this u-net implementation, there are multiple usages of "merge" along with "upsampling". For instance. up6 = merge([UpSampling2D(size=(2, 2))(conv5), after_conv4], mode='concat', concat_axis=1)

I can see that upsampling is to enlarge the size of layer, like conv5. In some paper, like Fully Convolutional Network, they usetransposed convolution, (the corresponding Keras implementation is Deconvolution2D). Does it make sense to use Deconvolution2D to replace upsampling2D?

Thanks.

EdwardTyantov commented 7 years ago

Hi, wenouyang. I found it not increasing any metrics at all, basically it's pretty much the same as upsampling + conv of the same size/stride.