There are a couple other questions that are similar, but I wanted to be specific.
Both this and this talk about adding additional data to a CNN by plugging into the lower layers of the CNN, or using it's output as the input to another network.
If the additional data in question maps to visual features in the image, wouldn't just stacking more channels on the input also work?
What would be the pros/cons of such an approach when adding 50 or 100 input channels on an existing network design like ResNet?
I ask because it seems like a much simpler approach than designing a custom architecture.