IIRC media players used to do something like a green screen where they draw a rectangle of a certain color and then a separate process would decode the video and display it over the colored rectangle. So taking a screenshot of the media player would capture the colored rectangle and it’d also be targeted by the video renderer.