Vector field functions

Initialization

affine_flow(→ torch.Tensor)	Compute dense flow field from homogeneous transformation.
compose_flows(→ torch.Tensor)	Compute composite flow field w = v o u = u(x) + v(x + u(x)).
zeros_flow(→ torch.Tensor)	Create batch of flow fields filled with zeros for given image batch size or grid.

Normalization

normalize_flow(→ torch.Tensor)	Map vectors with respect to unnormalized grid to vectors with respect to normalized grid.
denormalize_flow(→ torch.Tensor)	Map vectors with respect to normalized grid to vectors with respect to unnormalized grid.

Sampling

sample_flow(→ torch.Tensor)

Sample non-rigid flow fields at given points.

Warping

warp_grid(→ torch.Tensor)	Transform undeformed grid by a tensor of non-rigid flow fields.
warp_points(→ torch.Tensor)	Transform set of points by a tensor of non-rigid flow fields.
warp_image(→ torch.Tensor)	Sample data at optionally displaced grid points.

Miscellaneous

curl(→ torch.Tensor)	Calculate curl of vector field.
expv(→ torch.Tensor)	Group exponential maps of flow fields computed using scaling and squaring.
jacobian_det(→ torch.Tensor)	Evaluate Jacobian determinant of spatial deformation.