Rotscoping a Moving Object with Autotrack in SynthEyes

This tutorial shows an auto track of a shot with a moving object and camera. The camera is a tripod-type nodal motion, while a full 3-D solve is obtained for the moving object, and a coordinate system is set up and an object inserted into the coordinate frame of the moving object.

[@BrianHuether]

Hi, I am trying to figure out Syntheyes recommended workflow for my use case. I am filming myself to later insert in 3D scene in Unreal Engine. As experiment I filmed my cat. There is some camera movement, some cat movement. Ground is clearly in view. I am running Syntheyes on the original footage and have rotobrushed version of the footage such that background is transparent. I am just not sure about workflow. I see tutorials on using Make Grid the Ground. Other tutorials use three step manual coordinate setup process. Other tutorials have you draw cube then choose Make Mesh Ground. Can't figure out what would make sense for me.
Also, I am using GH5 camera where I knew focal length was 21 mm. Sensor is 17.3x13 but image is 1.77 crop. Not sure if I should enter camera details, as manual says maybe better to ignore.
Using demo version and trying to figure out workflow before buying.
Thanks!

[@borisfxlearn]

If you're trying to align the scene prior to export, take a little look at this video. It goes through reorienting and setting up correct scale : borisfx.com/videos/syntheyes-essentials-prepping-a-match-moved-scene-for-delivery/
What format are you taking into UE?
Normal workflow would be to do the tracking, create the 3D elements in UE, render those with the camera movement and do the final comp in another app. Sounds like you're using After Effects, so that will work.