Use the YOLOv5 model without additional training?

Question/Issue:

I’m working on a project to detect cars in images. I have a dataset of image that I captured that includes a car within each image.

As I’m working through the steps for the project, I notice that during the image labeling step, the suggested bounding boxes drawn using the YOLOv5 suggestion are very accurate, and correctly draw a bounding box around each car and label it as such.

Is it possible to simply use the YOLOv5 model to detect cars in images, without having to train it further on images I provide?

Is there a way to export the model, along with just a ‘car’ label for use in a project?

For context: The final model will be run on an OpenMV H7 Plus