Camera Animation Sequence Template Additive Issue

[LucasRidley]

Hi, I’m trying to create an animation library of camera mocap to be applied additively via Camera Anim.

The problem is I cannot get the Camera Anim to behave as additive to the sequencer camera motion.
The only way I’ve found for this to remotely work is to set the Hierarchical Bias of the to 0 of the Template Animation track that the camera anim lives on (pictured as idle breath for example).

My understanding is because the transform track inside the Camera Anim track is set to additive that this should already behave as an additive track once applied in sequencer, no?

Is there a better way to make a mocap library of camera animation to be additive on top of animation in sequencer?

One other issue I’m having is making the animation additive animation follow the camera, for example, if the additive animation includes translation down x axis and I rotate the camera 90 degrees to face the y-axis, the camera will still translate down the x-axis, instead of following this new orientation. Would there be a way to have additive animation that also follows the orientation of the camera it’s applied to?

I’m in 4.26.1

Thanks for any thoughts/help

[ludovicchabant]

Hi!

1. The solution is indeed to set HierarchicalBias to 0. In fact, one of the 4.26 patches should already make all Template Sequence sections default to that value. The reason is that, generally speaking, sub-sequences in Sequencer will “stomp” anything from the parent sequence(s) if they have a higher bias. If biases are the same, things get blended together instead of one stomping the other. This is how sub-scenes and shots work, for instance. Template sequences work the same.

2. Can you do what you want if no template sequences are involved? That is: can you achieve what you want with one single sequence that contains a transform track with 2 sections (one absolute, one additive)? If not, it wouldn’t work with a Template Sequence either.

[ludovicchabant]

1. Ah it looks like the fix to the bias default will be in 4.26.2.

2. I’m not quite sure what your use-case is If that helps, note that additive transform sections are applied in a channel-per-channel basis. That is: every channel gets added together and the transform is built only at the end. It’s not an “order of operation” thing where different additive sections get their transforms built and applied one after the other from top-most to bottom-most or something like that. So if you have an additive translation on X, it will move the object along X, whatever its orientation.

[LucasRidley]

Hey ludovicchabant, thanks for the response.

1. Makes sense for the first one, not sure why the defaults aren’t coming through on 4.26.1 to be 0 bias.
2. I’m not able to do it with two transform tracks. The path of motion always seems to be absolute/world space no matter what else I do on another transform track. It’s like a helicopter can keep flying straight while the cockpit spins beneath it, that’s what happens with adding a transform track. It seems like this should be possible, but maybe it’d an issue of an infinite loop, that the template can’t additively influence the camera and also follow it at the same time?