deniska wrote: ↑Wed Dec 18, 2019 4:30 pm
I am not sure that I agree with your horizon estimations.
Here is a simple drawing to illustrate how it appears to fail..
You are little bit high on your estimation of earth curvature (photo). I used earth curvature calculator:
- 2m eye level horizon distance (you can theoretically see object like coin at earth surface) is 5km, at 10km you can see 2m high object barely, 0.1m of the top.
- 10m jump camera level surface can be seen upto 11km
- 20m jump camera level you can see surface at 16km
If 1px/m is resolution at 10m jump height we would be thinking right angle triangle (for simplicity) of 11000 pixels wide and 10 pixels high. The hypotenuse would be practically parallel to 90 deg, exactly 0.05 degrees off ie down looking. If I did calculate correctly at 50m camera image would lie 5cm and at 1000m 90cm.
So it is pretty safe to say that the plane camera forms is pretty safe to 0-1km from camera. Over 1km you cannot see much of the jumper at wide angle action camera image. Quick calculation at drone gsd calculator at 1000m height gives 75cm/px with 2K video resolution, 1/2.3" sensor and 4mm focal length. So jumper would be 2-3 pixels at that 1000m distance already and camera lies 1 pixel - to fix you can move 3px kiter 1px up.
Maybe red kiter is max 50m away and yellow kiter max 200m? So it is pretty safe to say that scenario what you Deniska draw does not apply here.