Palettes Camera Comp Transform Matrix

 Hey guy's and gal's....I am looking for Ideas on how to animate the camera from the camera comp in the palette between transform matrices. I see where via python how we can save out the curent position via Transform Matrix and recall that Matrix  via python methods.

Screenshot 2021-04-29 142544

 When the saved transform matrices are recalled the camera "cuts" to the position recalled. I want the camera to animate or interpolate between those positions. Played around with a few ideas but haven't found anything that clicks...Anyone have an idea how to animate a camera between two transform matrices in the camera comp? Any ideas are appreciated.


Would a deck style approach with a camera blend work? Put the new transform matrix into the camera you’re not in, then cross into that camera.

I am working with a 3d scene from a Notch Block and the camera blend didn’t seem to work…but it is worth investigating further or at least asking Derivative. Thanks @raganmd

This a little awkward, but here’s another idea for you. Camera on the left is from the palette - this would be the camera you’d use to set up a shot.

You use an Object CHOP to capture the transform matrix of your interactive camera, then create “presets” by using snapping channels in a constant CHOP. You can then cross between presets (or you could use a blend CHOP for more complex transitions), which then get assigned to your output camera.

This is a bit messy, but it does give you an air gap between your interactive camera and your presets (which you may or may not want). I’m sure you could do this in the extension - but you’d need some code to handle the tween - which chops do nicely already.

Another version of this would be to write some python to handle storing presets that then just get written out to CHOPs to be used in an AB kind of style.

1 Like

@raganmd Thanks for the advice. I like this approach better than than using the transform matrix. It is simple and with a little python can make a cue bin and can also add a curve to the camera moves so they are not so linear. :wink:

1 Like

We’ve actually got an internal function that takes two cameras and a blend factor (0-1) and outputs a third camera that is the in between position. This method isn’t currently exposed in python, but I can look into making it available. It uses quaternions for the rotation blending, so it might get better results than the matrices.