how does VGG16 concat with the net that your owner defined ?

leeyeehoo / CSRNet-pytorch

CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes

642 stars 259 forks source link

Open linqiaozhou opened 5 years ago

linqiaozhou commented 5 years ago

I am confused in the code, how does the VGG16 concat with the net that your owner defined?

BedirYilmaz commented 5 years ago

There is no concatenation. The backend works on top of VGG-16, as an extension to the existing layers.

leeyeehoo commented 5 years ago

@BedirYilmaz he's right, I copy layers parameters from VGG to my networks

linqiaozhou commented 5 years ago

@BedirYilmaz he's right, I copy layers parameters from VGG to my networks

I see, thx~

wait1988 commented 5 years ago

Does using other backbone instead of VGG16 improve the performance?