Matching clip position and scale after warp

[Alan Lloyd]

Warp stabilizer alters the image, it may be subtle but it’s there. Maybe keyframe the motion effect on Clip 2 with uniform scaling turned off?

Another possible technique when doing that is to put frame 1 of clip 2 on track 2, and turn down the opacity while adjusting the scaling in order to match the two, then returning it to a match-frame and keyframing clip 2 back to “normal” over a few frames? That way, you’ll probably be able to see when you’re as close as possible to having identical frames.

Hope this helps.

[Alex White]

Hi Alan

I appreciate your response; I came to a similar conclusion last night (GMT).
Interesting, I’ve not had to do this before, situation hadn’t demanded it.
I have stabilized entire clips then transitioned to a different clip on the timeline but never attempted to stabilized part of a clip.
I can’t figure out why I’ve not had this in the past, this must be a real common problem.
“Hit record, reposition the camera several times during the shot only to find out in post that some of the repositioning was poor or jerky and attempt to fix it.

I have moved from shooting HD to 4k with the intent of using the scale and positioning in post to get my composition just right, then export in HD.
I feel that whilst a pan and tilt looks more aggressive when scale is increased it has shown this problem as a real issue.
Maybe I have overlooked something in my workflow.

Last night I attempted using two tracks like you suggested.
Original clip on V1 and a frozen frame from the same clip on V2 as a still image for position reference.
Both clips at identical scale.
I dropped the opacity on the still image on V2 and changed the blend just to make it stand out a little more.
Using a static object in the still frame (corner of a bookcase in this instance), I altered the position as I went from frame to frame in V1.
This worked out to be even worse, as you say, however subtle warp does change the image.

I also tried your suggestion of matching the frame with the last frame of the clip that had been stabilized, it still pops.

The best I can get is to use the pop and do something dramatic like punch in on the scale to make it look like an intentional digital zoom.
I.e. warp the first clip up to a practical point, cut and then change the scale in the next clip.
The real problem was because I was trying to keep the scale the same the pop was really noticeable.
By doing it this way at least it looks planned.

If only warp could do whatever magic is does under the hood and made the final frame match the last frame in the clip that would be so much better, as I say this surely is a common issue during a clip.

I suspect that people may recommend doing some kind of motion tracking in AE but it would be handy just to use the warp in PP for part of a clip.

If I am missing a real dumb piece of learning here, I’m more than happy for anyone to point it out to me.

Thanks