No, no copies are made at all.
After a video clip is uploaded to the cloud, the Person Detection model will be run against the uploaded video to check for a person. The model reads the video frames and gives the detection results for each frame. No copies are created, or saved, during this process.