So i’m trying to use the camera analysis node for the first time to get a camera solve for a 3d track. i have a lot of movement in the foreground and am trying to exclude some of it with an exclusion g mask. the issue i’m running into is that whenever i input the focal length/sensor size data in the camera analysis node, it massively offsets the exclusion mask i’ve loosely roto’d around one of the people going through the shot. is there any way to undo this offset? or a specific order of operations im supposed to follow to avoid this? when i check the gmask output, the mask is fine, but when i switch to the camera tracker media view, the mask has massively scaled up and offset from the subject.
How are you attaching your gmask to your camera analysis? I just tried this and I’m not seeing the gmask scale at all when I adjust focal length and sensor size.
Because the old Mono Analyzer didn’t like masks inside action, I always matte the footage before the action. Feed the action a clip with alpha maybe ¯\__(ツ)_/¯