-
Notifications
You must be signed in to change notification settings - Fork 21
Open
Description
Hello,
I would like to know if it can be possible to obtain for each animation the translation & scaling vectors + the quaternion instead of the final matrix4x4. Maybe we could create an enum
enum Transform {
// Either the final transform
Matrix(Matrix4<f32>),
// Or its decomposed form
LocRotScale {
t: Vector3<f32>,
r: Quaternion<f32>,
s: Vector3<f32>,
}
}
#[derive(Debug)]
pub struct Animation {
///
/// The node (joint) this animation is targeting
///
pub target: String,
///
/// Times for each sample (in seconds)
///
pub sample_times: Vec<f32>,
///
/// Node pose transforms for each sample.
/// Column-major.
///
pub sample_poses: Vec<Transform>,
}
I propose this because I think it is more easy to interpolate between keyframes if we have 2 LocRotScale than 2 Matrix4. I suppose there may be some problems when directly interpolating the coeffs of a Matrix4 or maybe I am wrong ?
best,
Metadata
Metadata
Assignees
Labels
No labels