Cal3DS2_Base - GTSAM Docs

Cal3DS2_Base is an internal class. It contains the shared pinhole and Brown–Conrady distortion operations used by Cal3DS2 and Cal3Unified; applications normally construct one of those concrete calibration models.

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

Obtaining the base interface¶

Although Python exposes a default Cal3DS2_Base() constructor, its zero focal lengths do not describe a usable camera. A concrete Cal3DS2 instance supplies valid parameters while exposing all inherited base methods.

calibration = gtsam.Cal3DS2(
    800.0, 790.0, 0.0, 320.0, 240.0,
    1e-3, -1e-5, 2e-4, -1e-4,
)
print("is a Cal3DS2_Base:", isinstance(calibration, gtsam.Cal3DS2_Base))

Shared parameter access¶

The base provides fx(), fy(), skew(), px(), py(), k1(), k2(), p1(), p2(), k(), K(), inverse(), and vector(). Manifold dimension and retraction belong to the concrete derived model.

print("intrinsic matrix:")
print(calibration.K())
print("distortion [k1, k2, p1, p2]:", calibration.k())
print("complete vector:", calibration.vector())

Shared pixel mapping¶

uncalibrate() applies radial and tangential distortion followed by the intrinsic matrix. calibrate() numerically inverts that transformation.

normalized = np.array([0.15, -0.08])
pixel = calibration.uncalibrate(normalized)
recovered = calibration.calibrate(pixel)
print("pixel:", pixel)
np.testing.assert_allclose(recovered, normalized, atol=1e-7)

Source¶

Cal3DS2_Base.h

AI assistance caveat¶

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