Question about resizing the input

[alirezanjf]

Hello and thank you for this great work. I am trying to train the network with the input images at a lower resolution, but I'm aware that I should changes radar-informed proposals as well and have a look at the annotations. Should I change them all manually? Is there a way, a method or anything that I can do this correctly with Detectron or any other ways. Can you please give me some suggestion on how to perform this task and what it is that I should keep in mind?