My problem is detecting object distance, size and maybe possible moving state. I am planning to make an app that is going to calculate these values based on photos taken and then narrate it to the user in speech. How should I approach to the main problem as a beginner, and what would you recommend as a tool/technology or algorithm to learn, what to do for the training stage?
I have seen various technologies such as Tensorflow, OpenCV, Keras and YOLO to use but I am still quite uncertain about learning and applying/integrating roadmap.