What is image data layout used in DNN models?

Question

RGB images can be stored in memory in different ways, e.g. see link below planar and interleaved: http://avitevet.com/uncategorized/when-to-use-it-interleaved-vs-planar-image-data-storage/

for example, for a RGB image:

Interleaved: Store all the bytes for pixel N before storing the bytes for pixel N + 1

Planar: Store all the bytes for color C before storing the bytes for color C + 1

In CNNs and DNNs which is the most common way of storing images? For example in CIFAR10, CIFAR100, and Imagenet datasets

In general it is only important for the the first layer, but I need to know that.