Actually I mostly love it - when it works it is a lifesaver, but I feel it’s buggy at times.
For the life of me I cannot figure out why a “patch” snaps into place sometimes and and sometimes does not. Can anyone enlighten me how I can get it to reliably work? Andy?
This is what I usually do:
1). Do a perspective track
2). Copy PG node, “Keep” one frame, invert, dangle from the original
3). Dangle a patch surface
Why won’t the patch snap into the exact position every time?
Also, PG doesn’t appear to give me a frame for the very last frame of the track range. Is this another case of “it’s not you, it’s me” - ?
I always check to see if my surface axis has been give a curious Z-depth.
The perspective grid works in 3D space and sometimes this effects how image surfaces are added into action.
I use it mainly for stabilise and re track work but I have found it a little buggy.
My safety net is always do a 3D perspective grid track. Copy that and make it 2D and then copy that again and use that as my invert for stabilising. Not often but sometimes when the grid looses its shit I can always recover it from the 2D saved version.
I have become more cavalier about how I stabilise and I feel comfortable adding an axis above to resize and reduce any scaling down or strange squashing of the stabiliser. It is easy to invert that later if your inverse axis sits under the re track perspective grid.
I love it. All of that planar tracking that came recently has been so helpful.
Mostly hate. I use the tracer masks a lot to planar track. I find the perspective grid doesn’t always look right. I can’t seem to bicubic when attached to perspective grid so have to use an extra action which gives a double aliasing.
Nah. To be honest I only saw someone tracking the perspective grid in 2D the other day. I thought it had to be tracked in 3D and then the 2D option was a conversion.
