Skip to main content

SO3/SE3 operations on any backend

Project description

nanomanifold

Fast, batched and differentiable SO3/SE3 transforms for any backend (NumPy, PyTorch, JAX, ...)

Works directly on arrays, defined as:

  • SO3: unit quaternions [w, x, y, z] for 3D rotations, shape (..., 4)
  • SE3: concatenated [quat, translation], shape (..., 7)
import numpy as np
from nanomanifold import SO3, SE3

# Rotations stored as quaternion arrays [w,x,y,z]
q = SO3.from_axis_angle(np.array([0, 0, 1]), np.pi/4)  # 45° around Z
points = np.array([[1, 0, 0], [0, 1, 0]])
rotated = SO3.rotate_points(q, points)

# Rigid transforms stored as 7D arrays [quat, translation]
T = SE3.from_rt(q, np.array([1, 0, 0]))  # rotation + translation
transformed = SE3.transform_points(T, points)

Installation

pip install nanomanifold

Quick Start

Rotations (SO3)

from nanomanifold import SO3

# Create rotations
q1 = SO3.from_axis_angle([1, 0, 0], np.pi/2)    # 90° around X
q2 = SO3.from_euler([0, 0, np.pi/4])            # 45° around Z
q3 = SO3.from_matrix(rotation_matrix)

# Compose and interpolate
q_combined = SO3.multiply(q1, q2)
q_halfway = SO3.slerp(q1, q2, t=0.5)

# Apply to points
points = np.array([[1, 0, 0], [0, 1, 0]])
rotated = SO3.rotate_points(q_combined, points)

Rigid Transforms (SE3)

from nanomanifold import SE3

# Create transforms
T1 = SE3.from_rt(q1, [1, 2, 3])               # rotation + translation
T2 = SE3.from_matrix(transformation_matrix)

# Compose transforms
T_combined = SE3.multiply(T1, T2)
T_inverse = SE3.inverse(T_combined)

# Apply to points
transformed = SE3.transform_points(T_combined, points)

API Reference

All functions are available via nanomanifold.SO3 and nanomanifold.SE3. Shapes follow the Array API convention and accept arbitrarily batched inputs.

SO3 (3D Rotations)

Function Signature
canonicalize(q) (...,4) -> (...,4)
to_axis_angle(q) (...,4) -> (...,3)
from_axis_angle(axis_angle) (...,3) -> (...,4)
to_euler(q, convention="ZYX") (...,4) -> (...,3)
from_euler(euler, convention="ZYX") (...,3) -> (...,4)
to_matrix(q) (...,4) -> (...,3,3)
from_matrix(R) (...,3,3) -> (...,4)
multiply(q1, q2) (...,4), (...,4) -> (...,4)
inverse(q) (...,4) -> (...,4)
rotate_points(q, points) (...,4), (...,N,3) -> (...,N,3)
slerp(q1, q2, t) (...,4), (...,4), (...,N) -> (...,N,4)
distance(q1, q2) (...,4), (...,4) -> (...)
log(q) (...,4) -> (...,3)
exp(tangent) (...,3) -> (...,4)
hat(w) (...,3) -> (...,3,3)
vee(W) (...,3,3) -> (...,3)
weighted_mean(quats, weights) sequence of (...,4), (...,N) -> (...,4)
mean(quats) sequence of (...,4) -> (...,4)

SE3 (Rigid Transforms)

Function Signature
canonicalize(se3) (...,7) -> (...,7)
from_rt(quat, translation) (...,4), (...,3) -> (...,7)
to_rt(se3) (...,7) -> (quat, translation)
from_matrix(T) (...,4,4) -> (...,7)
to_matrix(se3) (...,7) -> (...,4,4)
multiply(se3_1, se3_2) (...,7), (...,7) -> (...,7)
inverse(se3) (...,7) -> (...,7)
transform_points(se3, points) (...,7), (...,N,3) -> (...,N,3)
log(se3) (...,7) -> (...,6)
exp(tangent) (...,6) -> (...,7)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nanomanifold-0.1.2.tar.gz (76.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

nanomanifold-0.1.2-py3-none-any.whl (25.0 kB view details)

Uploaded Python 3

File details

Details for the file nanomanifold-0.1.2.tar.gz.

File metadata

  • Download URL: nanomanifold-0.1.2.tar.gz
  • Upload date:
  • Size: 76.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: uv/0.8.17

File hashes

Hashes for nanomanifold-0.1.2.tar.gz
Algorithm Hash digest
SHA256 11d03ed6d2192a6fc05deb6f53d7dabb3596fe73dd63c80bdeb5b5b8acbaab79
MD5 c93f722f46873b46db525a8bbf48139b
BLAKE2b-256 6c577cc03df7b3696b6056963486395c78e2fd80ac2429b8f4b62131f3694e98

See more details on using hashes here.

File details

Details for the file nanomanifold-0.1.2-py3-none-any.whl.

File metadata

File hashes

Hashes for nanomanifold-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 fe6a20d36effdad3ecff8b879df8127f12da1a4a80c37a4793cf5d73fee2be26
MD5 d7080072b4533da439a664d7a55a9d3b
BLAKE2b-256 7d17a4cf1b2e565320d7a2e5ca5f37765160d8c6fbe6d93925801965e780472d

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page