What is better for Object Recognition with Unity? OpenCV or YOLO?

I understand there are some GitHub links about integrating OpenCV and YOLO, but I was wondering what was better or just experiences in general?

Is there away to do all of this within Unity without doing some preliminary steps?

OpenCV supports YOLOv3 – YOLOv3 – Deep Learning Based Object Detection – YOLOv3 with OpenCV ( Python / C++ )