pytorch3d.renderer.implicit.utils

utils

class pytorch3d.renderer.implicit.utils.RayBundle(origins: Tensor, directions: Tensor, lengths: Tensor, xys: Tensor)

Bases: NamedTuple

Parametrizes points along projection rays by storing:

origins: A tensor of shape (…, 3) denoting the

origins of the sampling rays in world coords.

directions: A tensor of shape (…, 3) containing the direction

vectors of sampling rays in world coords. They don’t have to be normalized; they define unit vectors in the respective 1D coordinate systems; see documentation for ray_bundle_to_ray_points() for the conversion formula.

lengths: A tensor of shape (…, num_points_per_ray)

containing the lengths at which the rays are sampled.

xys: A tensor of shape (…, 2), the xy-locations (xys) of the ray pixels

origins: Tensor

Alias for field number 0

directions: Tensor

Alias for field number 1

lengths: Tensor

Alias for field number 2

xys: Tensor

Alias for field number 3

class pytorch3d.renderer.implicit.utils.HeterogeneousRayBundle(origins: Tensor, directions: Tensor, lengths: Tensor, xys: Tensor, camera_ids: LongTensor | None = None, camera_counts: LongTensor | None = None)

Bases: object

Members:

origins: A tensor of shape (…, 3) denoting the

origins of the sampling rays in world coords.

directions: A tensor of shape (…, 3) containing the direction

vectors of sampling rays in world coords. They don’t have to be normalized; they define unit vectors in the respective 1D coordinate systems; see documentation for ray_bundle_to_ray_points() for the conversion formula.

lengths: A tensor of shape (…, num_points_per_ray)

containing the lengths at which the rays are sampled.

xys: A tensor of shape (…, 2), the xy-locations (xys) of the ray pixels camera_ids: A tensor of shape (N, ) which indicates which camera

was used to sample the rays. N is the number of unique sampled cameras.

camera_counts: A tensor of shape (N, ) which how many times the

coresponding camera in camera_ids was sampled. sum(camera_counts)==total_number_of_rays

If we sample cameras of ids [0, 3, 5, 3, 1, 0, 0] that would be stored as camera_ids=[1, 3, 5, 0] and camera_counts=[1, 2, 1, 3]. camera_ids is a set like object with no particular ordering of elements. ith element of camera_ids coresponds to the ith element of camera_counts.

origins: Tensor

directions: Tensor

lengths: Tensor

xys: Tensor

camera_ids: LongTensor | None = None

camera_counts: LongTensor | None = None

pytorch3d.renderer.implicit.utils.ray_bundle_to_ray_points(ray_bundle: RayBundle | HeterogeneousRayBundle) → Tensor

Converts rays parametrized with a ray_bundle (an instance of the RayBundle named tuple or HeterogeneousRayBundle dataclass) to 3D points by extending each ray according to the corresponding length.

E.g. for 2 dimensional tensors ray_bundle.origins, ray_bundle.directions

and ray_bundle.lengths, the ray point at position [i, j] is:

ray_bundle.points[i, j, :] = (
    ray_bundle.origins[i, :]
    + ray_bundle.directions[i, :] * ray_bundle.lengths[i, j]
)

Note that both the directions and magnitudes of the vectors in ray_bundle.directions matter.

Parameters:

ray_bundle – A RayBundle or HeterogeneousRayBundle object with fields: origins: A tensor of shape (…, 3) directions: A tensor of shape (…, 3) lengths: A tensor of shape (…, num_points_per_ray)

Returns:

rays_points –

A tensor of shape (…, num_points_per_ray, 3)

containing the points sampled along each ray.

pytorch3d.renderer.implicit.utils.ray_bundle_variables_to_ray_points(rays_origins: Tensor, rays_directions: Tensor, rays_lengths: Tensor) → Tensor

Converts rays parametrized with origins and directions to 3D points by extending each ray according to the corresponding ray length:

E.g. for 2 dimensional input tensors rays_origins, rays_directions and rays_lengths, the ray point at position [i, j] is:

rays_points[i, j, :] = (
    rays_origins[i, :]
    + rays_directions[i, :] * rays_lengths[i, j]
)

Note that both the directions and magnitudes of the vectors in rays_directions matter.

Parameters:

• rays_origins – A tensor of shape (…, 3)

• rays_directions – A tensor of shape (…, 3)

• rays_lengths – A tensor of shape (…, num_points_per_ray)

Returns:

rays_points –

A tensor of shape (…, num_points_per_ray, 3)

containing the points sampled along each ray.