Thank you for reading this far
I was wondering if it is possible to make some kind of machine learning model that made an inference on each of the horizontal lines of an image. That is if I have a 1920x1080 image where each of the 1920 lines represents a spectra and outputs 1080 predictions. I belive that it could be beneficial for the training if I could input this prior knowledge into the model.
I want to be able to do this to port it by high-level synthesis to an FPGA for effective edge computing and hoped to be able to generate one (1) TensorFlow model for effective edge computing. The idea was to use PYNQ.
If anyone anywhere has anything to add to this idea, e.g. it is a bad idea, I would love to hear from you!