peterWon / kontiki

Toolkit for continuous-time structure from motion

Kontiki - the continuous time toolkit

Kontiki is a toolkit for continuous-time structure from motion. In short, it can estimate a trajectory (and 3D structure) from a set of measurements.

The documentation is available at https://hovren.github.io/kontiki/.

Example: Visual-inertial structure from motion

import numpy as np
import itertools
from kontiki import trajectories, sensors, TrajectoryEstimator
import kontiki.measurements as M

views, landmarks = load_my_sfm_data()

trajectory = trajectories.UniformSE3SplineTrajectory()
trajectory.extend_to(some_max_time, np.eye(4))

imu = sensors.BasicImu()

camera = sensors.PinholeCamera(720, 1280, 0.03, my_camera_matrix)

gyro_meas = [M.GyroscopeMeasurement(imu, t, x)
             for t, x in my_gyro_data]
acc_meas = [M.AccelerometerMeasurement(imu, t, x)
            for t, x in my_acc_data]
im_meas = [M.StaticRsCameraMeasurement(camera, obs)
           for v in view for obs in v.observations]

estimator = TrajectoryEstimator(trajectory)
for m in itertools.chain(gyro_meas, acc_meas, im_meas):
    estimator.add_measurement(m)

summary = estimator.solve()

Installation

See the documentation at https://hovren.github.io/kontiki/users/install.html.