You do not have a good understanding of what Camera Tracking does or how to align layers in 3D space after the camera tracking has created a new camera. Your last screenshot shows you have made a copy of the footage layer 3D. That won't work at all. The footage has to be 2D, the text layer must either be added to a surface or you have to Shift + parent the 3D text layer to a reference solid or null added to the comp that has been attached to a surface. I don't see anything in your workflow that indicates you have spent any time looking at the User Guide or a good tutorial on Camera Tracking.
To have camera tracking come up with a good solution for a camera you have to have a fair amount of fixed geometry in a scene that has flat surfaces that do not move. Sidewalks, streets, Buildings, Streetlights, signs are good. Widows, because they have reflections that change when the camera moves, people that are walking around, anything that the wind is blowing like a wheat field, water with ripples, moving traffic and anything that does not have any detail like the overexposed side of a white building are bad things to have in a shot you expect to camera track.
When you run camera tracking, if there is something in the shot that represents the ground you walk on then you should set an origin and ground plane, then attach a reference solid and add a camera to that spot using exactly the same tracking markers to make sure that the camera solution is accurate. If the camera is on a tripod and panning you will get a camera pan solution and that has no depth information so you cannot set a ground plane and origin for that kind of shot.
Once you set an origin and ground plane and add a solid in the same spot it is important that you check the track for accuracy. When that is done you can add other solids to use as references or text layers or even nulls but you must be very careful that the tracking markers you pick are aligned on the same surface (plane). I never use Nulls because you can't easily see if they stick to the surface, but I do use solids and often add grids to them to make sure that the track is accurate.
Once you have the surfaces identified and have made sure you picked the right surfaces (tracking markers) it is time to start adding the 3D elements you intend to put in your scene. You can easily snap any 3D layer to the same position and orientation as a reference layer, one of the track solids, by holding down the shift key and parenting the new 3D layer to the selected track solid. Then you can carefully adjust the rotation values to square the layer up to the surface.
The origin and ground plane you establish will be at 0, 0, not at comp center so using the other views to line things up in AE's 3D world is a bit tricky, but you can adjust the custom, top, left, right, front, back and bottom views with the camera tool (c) to make things easier.
You cannot add another camera and have it work. You cannot just drop in a 3D layer and expect it to be in the right position. If you have something you want to appear to be in the 3D scene but there is no surface that you can use to attach the new layer to you'll have to set up two or three reference layers and then move the new layer into position by carefully manipulating the x, y and z values so that the new 3D layer properly lines up.
Most of the camera tracking tutorials that I see are presented by enthusiasts and they leave out a bunch of critical information and promote pretty sloppy and inefficient workflows. I would start learning Camera Tracking using the Learn resources right on the Home Screen in AE, in the Resources section of this forum under User Guide or on the After Effects product page. They will take you here: Learn AE. Jus poking around on YouTube or fiddling with AE will just cause you a lot of frustration. Carefully study the tutorials Adobe has saved in the User Guide section and spend some time reading and you'll get it figured out.
1 Correct answer
6
Replies
AdChoices