An extension of separable convolutions for 3D volumes. Performs volumetric convolutions for each channel of the input volume and will increase the output volume based on the number of convolutional operations (denoted as depth_multiplier
inside the code)
Base code for the implementation is used from: https://github.com/titu1994/MobileNetworks/blob/master/depthwise_conv.py
- Keras 2.2.4+
- Tensorflow 1.13
from DepthwiseConv3D import DepthwiseConv3D
input = Input(...)
x = DepthwiseConv3D(kernel_size=(3,3,3), depth_multiplier=2)(input)
...
[1] F. Chollet, Xception: Deep Learning with Depthwise Separable Convolutions [link]