SmartFactorBase - GTSAM Docs

The SmartFactorBase class is an internal base class for wrapped smart projection factors. It provides their shared measurement and camera workflow.

Authors: Frank Dellaert, et al. (see THANKS for the full author list)

See LICENSE for the license information

try:
    import google.colab
    %pip install --quiet gtsam-develop
except ImportError:
    pass

import gtsam
import numpy as np

from gtsam.symbol_shorthand import X

Use the base API through a concrete smart factor¶

SmartFactorBase is not constructed directly. We use SmartProjectionFactorPinholeCameraCal3_S2, which inherits the base API, and add two image measurements of one landmark. The base class stores the measurement vector and the corresponding camera keys while leaving the landmark implicit.

calibration = gtsam.Cal3_S2(500.0, 500.0, 0.0, 320.0, 240.0)
cameras = [
    gtsam.PinholeCameraCal3_S2(gtsam.Pose3(), calibration),
    gtsam.PinholeCameraCal3_S2(
        gtsam.Pose3(gtsam.Rot3(), np.array([1.0, 0.0, 0.0])),
        calibration,
    ),
]
landmark = np.array([0.0, 0.0, 5.0])
factor = gtsam.SmartProjectionFactorPinholeCameraCal3_S2(
    gtsam.noiseModel.Isotropic.Sigma(2, 1.0)
)
values = gtsam.Values()
for index, camera in enumerate(cameras):
    factor.add(camera.project(landmark), X(index))
    values.insert(X(index), camera)

assert len(factor.measured()) == 2
print("measurements:", factor.measured())
print("camera keys:", factor.keys())

measurements: [array([320., 240.]), array([220., 240.])]
camera keys: [8646911284551352320, 8646911284551352321]

Recover the cameras and triangulate¶

cameras(values) follows the stored keys and assembles the typed camera set used by the derived factor. With consistent measurements, the factor triangulates the landmark and reports zero total reprojection error.

camera_set = factor.cameras(values)
assert len(camera_set) == 2
assert np.isclose(factor.totalReprojectionError(camera_set), 0.0)
print("camera set type:", type(camera_set).__name__)
print("triangulation valid:", bool(factor.point(values)))

camera set type: CameraSetCal3_S2
triangulation valid: True

When to use it¶

SmartFactorBase is internal shared machinery for smart projection factors. Application code should construct a derived factor matching the camera type stored in Values; the wrapper supplies the corresponding base specialization automatically. The base API is most useful when adding measurements or inspecting how a smart factor resolves its cameras.

Continue with SmartProjectionFactor for the nonlinear factor, SmartFactorParams for configuration, and JacobianFactorQ for one possible linearization result.

Source¶

SmartFactorBase.h

AI assistance caveat¶

AI was used to help draft this documentation, and inaccuracies could be present.