I am unsure about how to word this question correctly, edits appreciated.
I am trying to create a neural network model that can predict distance from camera feed. And I am doing it by feeding actual distance data from an images corresponding LIDAR and RADAR data.
Is this doable?
I am new to depth perception and distance estimation. What I am trying to achieve is similar to a methodology described here Basically, take an image, lable the objects or make use of an object recognition model like YOLOnet for that and then give the distance of the said object from the camera as training data. Set only the corresponding distance of the objects in the images as ground truth data.
I am unsure of how I should go about implementing it. Am I doing anything wrong with the described methodology or is this something not possible?
