StereoPoint2

StereoPoint2 stores a rectified stereo measurement (uL, uR, v): horizontal coordinates in the left and right images plus their shared vertical coordinate.

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

Initialization¶

Construct from three scalars or a three-vector. Identity() returns the zero measurement.

measurement = gtsam.StereoPoint2(340.0, 315.0, 205.0)
from_vector = gtsam.StereoPoint2(np.array([340.0, 315.0, 205.0]))
print("same measurement:", measurement.equals(from_vector, 1e-12))

Components and disparity¶

uL(), uR(), and v() expose the components, while vector() returns them in constructor order. The horizontal disparity uL - uR is inversely related to depth for a rectified stereo pair.

disparity = measurement.uL() - measurement.uR()
print("measurement vector:", measurement.vector())
print("disparity:", disparity)
assert disparity > 0.0

Use with `StereoCamera`¶

StereoCamera.project() produces this type, and StereoCamera.backproject() consumes it. Keeping the shared v coordinate explicit encodes the rectified-image assumption.

The value type does not enforce measurement validity. In normal rectified geometry, positive finite depth produces positive disparity; zero disparity corresponds to a point at infinity, and negative disparity usually indicates a mismatched observation or inconsistent camera convention.

Source¶

StereoPoint2.h

AI assistance caveat¶

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

Initialization¶

Components and disparity¶

Use with StereoCamera¶

Source¶

AI assistance caveat¶

Use with `StereoCamera`¶