yes. put the plane outside of the tracker and focal distance, but as a child of the camera, and assign it’s position, rotation and scale manually using the values from the tracker and freeze their values instead of texture using delay value manually or use patch called “store vec3” or something along that line.
Adi is right about using delay value, but you could also use a scene render pass to render all of the elements (including the background) into a texture, then feed that texture into the freeze frame graph (instead of the camera texture)
Thank you so so much guys
I tried @josh_beckwith’s method and it worked but the problem with it was that I couldn’t control the visibility of the plane after it’s being frozen but @Adi_Satrio’s method worked for me as I expected. here is the working node for whoever that is interested.
can we just have a 3D object as child of face tracker and still freeze it with store vec3 (not renderpass)? cause it’s much more accurate than manualy assigning position, rotation to something outside facetracker
no, because the rotation and origin of the parents will still affect the local transform. it has to be in that exact hierarchy to get the 1:1 transformation that you get from the face tracker patch. try it yourself. you have 2 things to capture/store. the 3d object transform (in this case the plane) and the background texture (in this case the camera texture). you can’t just freeze one of them and expect them to freeze in place, another way would be by simply freeze the final render output before the device output node. it’s mainly for post processing effect. there are pros and cons between each method, you’ll have to decide which one that suits your need.
the downside of freezing the final render output, it means any other objects or design element will also be static. that’s a problem if you want to just freeze 1 object and the background but maybe you want after they freeze there are another element that still moving like particles, 3d object, animation sequence. the simple example would be something like the old school Thug Life meme, where the video would be stopped and then a sunglasses and the tagline text “Thug Life” appear and goes to the last eye and face position. or maybe for meme / funny fail effect where people will stop the video when someone or something is about to fall or fail, and some other stuffs appears like “to be continued…” or something.