According to my previous question in ventral Stream pathway and architecture, I want now to get a brief example about how the S1 layer is constructed. In other words, how all the simple units are tuned with the gaussian-Like tuning (for example). I am only interested to get such a cartoon example in step by step how this operation can be achieved in given of inputs (which inputs? and what we mean about these inputs?) in order to obtain the tuned simple units.
We all know that each simple unit is obtained after a tuning operation around their inputs (Subunits) in order to select an optimal output which corresponds to the preferred orientation for such a simple unit.
Moreover, I know that the S1 layer units perform a convolution on regions of the raw input image using gabor filters at different orientations and sizes. The entire population of S1 units represents a convolution map of Gabor filters of different sizes and orientations with the entire raw image (really I didn't understand this point) you can read the subsection 2.2 in this article
The image below contains the image to be recognized by the brain and some simple units obtained after the tuning operation. So what I want is to obtain a brief example which can include the details (step by step) of how the operation of tuning can be achieved.
i didn't understand well the concept. That's why i need a real example with a specific image which can contains all the steps described in my attached image (step by step) because i still don't understand what we mean about inputs x, etc. So please if anyone can give me a real example with a specific image which can respond briefly to this attached image